10-backtest-risk-parity.ipynb

{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Risk Parity backtest\n",
    "\n",
    "This notebook follows the 06a-backtest example and implements risk-parity weighting in addition to equal-weighting."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: pymssql in c:\\users\\kpc2\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (2.2.7)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "[notice] A new release of pip available: 22.2.1 -> 23.0.1\n",
      "[notice] To update, run: python.exe -m pip install --upgrade pip\n",
      "\n",
      "[notice] A new release of pip available: 22.2.1 -> 23.0.1\n",
      "[notice] To update, run: python.exe -m pip install --upgrade pip\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: scikit-learn in c:\\users\\kpc2\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (1.2.2)\n",
      "Requirement already satisfied: scipy>=1.3.2 in c:\\users\\kpc2\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from scikit-learn) (1.9.0)\n",
      "Requirement already satisfied: numpy>=1.17.3 in c:\\users\\kpc2\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from scikit-learn) (1.23.1)\n",
      "Requirement already satisfied: threadpoolctl>=2.0.0 in c:\\users\\kpc2\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from scikit-learn) (3.1.0)\n",
      "Requirement already satisfied: joblib>=1.1.1 in c:\\users\\kpc2\\appdata\\local\\programs\\python\\python310\\lib\\site-packages (from scikit-learn) (1.2.0)\n"
     ]
    }
   ],
   "source": [
    "from sqlalchemy import create_engine\n",
    "!pip install pymssql\n",
    "!pip install scikit-learn\n",
    "import pymssql\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.preprocessing import QuantileTransformer\n",
    "from sklearn.preprocessing import PolynomialFeatures\n",
    "from sklearn.pipeline import make_pipeline\n",
    "\n",
    "model = LinearRegression()\n",
    "qt = QuantileTransformer(output_distribution=\"normal\")\n",
    "poly = PolynomialFeatures(degree=2)\n",
    "pipe = make_pipeline(poly, model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "server = \"mssql-82792-0.cloudclusters.net:16272\"\n",
    "username = \"user\"\n",
    "password = \"RiceOwls1912\" \n",
    "dfbase = \"ghz\"\n",
    "string = \"mssql+pymssql://\" + username + \":\" + password + \"@\" + server + \"/\" + dfbase\n",
    "conn = create_engine(string).connect()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Get data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_sql(\n",
    "    \"\"\"\n",
    "    select ticker, date, ret, roeq, bm, mom12m, mve\n",
    "    from data\n",
    "    where date>='2000-01'\n",
    "    order by date, ticker\n",
    "    \"\"\", \n",
    "    conn\n",
    ")\n",
    "df = df.dropna()\n",
    "conn.close()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "raw_data = df.copy()\n",
    "raw_data = raw_data.reset_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "features = [\"roeq\", \"bm\", \"mom12m\"]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>ret</th>\n",
       "      <th>roeq</th>\n",
       "      <th>bm</th>\n",
       "      <th>mom12m</th>\n",
       "      <th>mve</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>date</th>\n",
       "      <th>ticker</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"11\" valign=\"top\">2000-01</th>\n",
       "      <th>AA</th>\n",
       "      <td>-0.160392</td>\n",
       "      <td>0.039969</td>\n",
       "      <td>0.438760</td>\n",
       "      <td>0.783092</td>\n",
       "      <td>17.230341</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AABC</th>\n",
       "      <td>-0.029528</td>\n",
       "      <td>0.054207</td>\n",
       "      <td>1.107577</td>\n",
       "      <td>0.077586</td>\n",
       "      <td>9.194465</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAC</th>\n",
       "      <td>0.056338</td>\n",
       "      <td>0.053986</td>\n",
       "      <td>1.893766</td>\n",
       "      <td>0.241379</td>\n",
       "      <td>12.072728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AACE</th>\n",
       "      <td>-0.067568</td>\n",
       "      <td>0.049065</td>\n",
       "      <td>0.339860</td>\n",
       "      <td>0.095833</td>\n",
       "      <td>12.134491</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAG</th>\n",
       "      <td>-0.100694</td>\n",
       "      <td>0.028019</td>\n",
       "      <td>0.703278</td>\n",
       "      <td>-0.294565</td>\n",
       "      <td>13.544615</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ZOMX</th>\n",
       "      <td>-0.005525</td>\n",
       "      <td>0.094583</td>\n",
       "      <td>0.428749</td>\n",
       "      <td>3.557692</td>\n",
       "      <td>13.459701</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ZONA</th>\n",
       "      <td>0.771429</td>\n",
       "      <td>-0.078612</td>\n",
       "      <td>0.249106</td>\n",
       "      <td>-0.715686</td>\n",
       "      <td>10.805451</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ZQK</th>\n",
       "      <td>-0.217742</td>\n",
       "      <td>0.041841</td>\n",
       "      <td>0.397946</td>\n",
       "      <td>-0.243750</td>\n",
       "      <td>12.770872</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ZRAN</th>\n",
       "      <td>-0.181614</td>\n",
       "      <td>0.033529</td>\n",
       "      <td>0.202465</td>\n",
       "      <td>1.278572</td>\n",
       "      <td>13.572464</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ZVXI</th>\n",
       "      <td>0.926829</td>\n",
       "      <td>0.014941</td>\n",
       "      <td>1.241130</td>\n",
       "      <td>0.052632</td>\n",
       "      <td>9.771819</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5063 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                     ret      roeq        bm    mom12m        mve\n",
       "date    ticker                                                   \n",
       "2000-01 AA     -0.160392  0.039969  0.438760  0.783092  17.230341\n",
       "        AABC   -0.029528  0.054207  1.107577  0.077586   9.194465\n",
       "        AAC     0.056338  0.053986  1.893766  0.241379  12.072728\n",
       "        AACE   -0.067568  0.049065  0.339860  0.095833  12.134491\n",
       "        AAG    -0.100694  0.028019  0.703278 -0.294565  13.544615\n",
       "...                  ...       ...       ...       ...        ...\n",
       "        ZOMX   -0.005525  0.094583  0.428749  3.557692  13.459701\n",
       "        ZONA    0.771429 -0.078612  0.249106 -0.715686  10.805451\n",
       "        ZQK    -0.217742  0.041841  0.397946 -0.243750  12.770872\n",
       "        ZRAN   -0.181614  0.033529  0.202465  1.278572  13.572464\n",
       "        ZVXI    0.926829  0.014941  1.241130  0.052632   9.771819\n",
       "\n",
       "[5063 rows x 5 columns]"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "raw_data.xs('IBM', level='ticker', drop_level=False)\n",
    "raw_data.xs('2000-01', level='date', drop_level=False)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Estimate expanding window volatilities each month"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [],
   "source": [
    "WINDOW = 60"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "ticker  date   \n",
       "A       2007-03    0.128052\n",
       "        2007-04    0.126989\n",
       "        2007-05    0.126610\n",
       "        2007-06    0.125585\n",
       "        2007-07    0.124604\n",
       "                     ...   \n",
       "ZVXI    2021-11    0.273161\n",
       "        2021-12    0.273161\n",
       "        2022-01    0.273161\n",
       "        2022-02    0.273161\n",
       "        2022-03    0.273161\n",
       "Name: sd, Length: 928365, dtype: float64"
      ]
     },
     "execution_count": 95,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = raw_data.set_index(['ticker','date']).sort_values(by=['ticker','date'])\n",
    "df_wide = df.ret.unstack(level=0)\n",
    "sds = df_wide.expanding(WINDOW).std()\n",
    "sds = sds.stack()\n",
    "sds = sds.swaplevel().sort_index()\n",
    "sds.name = 'sd'\n",
    "sds"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "ticker  date   \n",
       "IBM     2004-12    0.103784\n",
       "        2005-01    0.103169\n",
       "        2005-02    0.102328\n",
       "        2005-03    0.101519\n",
       "        2005-04    0.102848\n",
       "                     ...   \n",
       "        2021-11    0.071878\n",
       "        2021-12    0.072231\n",
       "        2022-01    0.072095\n",
       "        2022-02    0.072109\n",
       "        2022-03    0.072056\n",
       "Name: sd, Length: 208, dtype: float64"
      ]
     },
     "execution_count": 96,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sds.xs('IBM',level='ticker', drop_level=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Drop largest 500 stocks each month"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = raw_data.set_index([\"date\", \"ticker\"]).copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {},
   "outputs": [],
   "source": [
    "df[\"size_rnk\"] = df.groupby(\"date\").mve.rank(ascending=False)\n",
    "df = df[df.size_rnk>500]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Transform features each month"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {},
   "outputs": [],
   "source": [
    "def qt_df(d):\n",
    "    x = qt.fit_transform(d)\n",
    "    return pd.DataFrame(x, columns=d.columns, index=d.index)\n",
    "\n",
    "df[features] = df.groupby(\"date\", group_keys=False)[features].apply(qt_df)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Transform target each month"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {},
   "outputs": [],
   "source": [
    "def qt_ser(s):\n",
    "    x = s.copy()\n",
    "    x = x.to_numpy().reshape(-1, 1)\n",
    "    x = qt.fit_transform(x).flatten()\n",
    "    return pd.Series(x, index=s.index)\n",
    "\n",
    "df[\"target\"] = df.groupby(\"date\", group_keys=False).ret.apply(qt_ser)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Train and predict in a loop"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {},
   "outputs": [],
   "source": [
    "predictions = None\n",
    "\n",
    "dates = [\"2005-01\", \"2010-01\", \"2015-01\", \"2020-01\", \"3000-01\"]\n",
    "for train_date, end_date in zip(dates[:-1], dates[1:]):\n",
    "\n",
    "    filter1 = df.index.get_level_values(\"date\") < train_date\n",
    "    filter2 = df.index.get_level_values(\"date\") < end_date\n",
    "\n",
    "    train = df[filter1]\n",
    "    test = df[~filter1 & filter2]\n",
    "\n",
    "    Xtrain = train[features]\n",
    "    ytrain = train[\"target\"]\n",
    "    Xtest = test[features]\n",
    "\n",
    "    pipe.fit(Xtrain, ytrain)\n",
    "    pred = pipe.predict(Xtest)\n",
    "    pred = pd.Series(pred, index=test.index)\n",
    "    predictions = pd.concat((predictions, pred))\n",
    "\n",
    "df[\"predict\"] = predictions"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Merge in volatilities"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>ret</th>\n",
       "      <th>roeq</th>\n",
       "      <th>bm</th>\n",
       "      <th>mom12m</th>\n",
       "      <th>mve</th>\n",
       "      <th>size_rnk</th>\n",
       "      <th>target</th>\n",
       "      <th>predict</th>\n",
       "      <th>sd</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>date</th>\n",
       "      <th>ticker</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"5\" valign=\"top\">2004-12</th>\n",
       "      <th>AACE</th>\n",
       "      <td>0.111903</td>\n",
       "      <td>0.012259</td>\n",
       "      <td>0.686011</td>\n",
       "      <td>0.846174</td>\n",
       "      <td>12.790238</td>\n",
       "      <td>2047.0</td>\n",
       "      <td>0.725076</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.116874</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAII</th>\n",
       "      <td>0.051780</td>\n",
       "      <td>2.097185</td>\n",
       "      <td>-1.435987</td>\n",
       "      <td>-2.726759</td>\n",
       "      <td>11.388843</td>\n",
       "      <td>3170.0</td>\n",
       "      <td>0.222377</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.245369</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAME</th>\n",
       "      <td>0.033333</td>\n",
       "      <td>-0.282913</td>\n",
       "      <td>1.728346</td>\n",
       "      <td>0.298846</td>\n",
       "      <td>11.058055</td>\n",
       "      <td>3391.0</td>\n",
       "      <td>0.003373</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.121825</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AANB</th>\n",
       "      <td>0.180442</td>\n",
       "      <td>0.505941</td>\n",
       "      <td>-0.247464</td>\n",
       "      <td>0.093628</td>\n",
       "      <td>10.909255</td>\n",
       "      <td>3491.0</td>\n",
       "      <td>1.162363</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.079512</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AAON</th>\n",
       "      <td>0.009422</td>\n",
       "      <td>0.061443</td>\n",
       "      <td>-0.674471</td>\n",
       "      <td>-0.720448</td>\n",
       "      <td>12.193995</td>\n",
       "      <td>2534.0</td>\n",
       "      <td>-0.342627</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.113154</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"5\" valign=\"top\">2022-03</th>\n",
       "      <th>YELP</th>\n",
       "      <td>0.006492</td>\n",
       "      <td>-0.008022</td>\n",
       "      <td>-0.296959</td>\n",
       "      <td>-0.173110</td>\n",
       "      <td>14.721554</td>\n",
       "      <td>1002.0</td>\n",
       "      <td>0.143844</td>\n",
       "      <td>0.036309</td>\n",
       "      <td>0.147324</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>YORW</th>\n",
       "      <td>0.002229</td>\n",
       "      <td>0.344067</td>\n",
       "      <td>-0.624169</td>\n",
       "      <td>0.384817</td>\n",
       "      <td>13.285281</td>\n",
       "      <td>1535.0</td>\n",
       "      <td>0.091432</td>\n",
       "      <td>0.056389</td>\n",
       "      <td>0.061721</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>YTEN</th>\n",
       "      <td>0.261905</td>\n",
       "      <td>-1.262128</td>\n",
       "      <td>0.071652</td>\n",
       "      <td>-1.837949</td>\n",
       "      <td>9.928395</td>\n",
       "      <td>2311.0</td>\n",
       "      <td>1.597743</td>\n",
       "      <td>-0.169210</td>\n",
       "      <td>0.310849</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ZEUS</th>\n",
       "      <td>0.436683</td>\n",
       "      <td>1.467827</td>\n",
       "      <td>1.924874</td>\n",
       "      <td>0.666564</td>\n",
       "      <td>12.604142</td>\n",
       "      <td>1728.0</td>\n",
       "      <td>2.168410</td>\n",
       "      <td>0.130236</td>\n",
       "      <td>0.178060</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ZUMZ</th>\n",
       "      <td>-0.141155</td>\n",
       "      <td>0.598113</td>\n",
       "      <td>0.083015</td>\n",
       "      <td>0.054865</td>\n",
       "      <td>13.757728</td>\n",
       "      <td>1370.0</td>\n",
       "      <td>-1.378972</td>\n",
       "      <td>0.072304</td>\n",
       "      <td>0.157737</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>430757 rows × 9 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                     ret      roeq        bm    mom12m        mve  size_rnk  \\\n",
       "date    ticker                                                                \n",
       "2004-12 AACE    0.111903  0.012259  0.686011  0.846174  12.790238    2047.0   \n",
       "        AAII    0.051780  2.097185 -1.435987 -2.726759  11.388843    3170.0   \n",
       "        AAME    0.033333 -0.282913  1.728346  0.298846  11.058055    3391.0   \n",
       "        AANB    0.180442  0.505941 -0.247464  0.093628  10.909255    3491.0   \n",
       "        AAON    0.009422  0.061443 -0.674471 -0.720448  12.193995    2534.0   \n",
       "...                  ...       ...       ...       ...        ...       ...   \n",
       "2022-03 YELP    0.006492 -0.008022 -0.296959 -0.173110  14.721554    1002.0   \n",
       "        YORW    0.002229  0.344067 -0.624169  0.384817  13.285281    1535.0   \n",
       "        YTEN    0.261905 -1.262128  0.071652 -1.837949   9.928395    2311.0   \n",
       "        ZEUS    0.436683  1.467827  1.924874  0.666564  12.604142    1728.0   \n",
       "        ZUMZ   -0.141155  0.598113  0.083015  0.054865  13.757728    1370.0   \n",
       "\n",
       "                  target   predict        sd  \n",
       "date    ticker                                \n",
       "2004-12 AACE    0.725076       NaN  0.116874  \n",
       "        AAII    0.222377       NaN  0.245369  \n",
       "        AAME    0.003373       NaN  0.121825  \n",
       "        AANB    1.162363       NaN  0.079512  \n",
       "        AAON   -0.342627       NaN  0.113154  \n",
       "...                  ...       ...       ...  \n",
       "2022-03 YELP    0.143844  0.036309  0.147324  \n",
       "        YORW    0.091432  0.056389  0.061721  \n",
       "        YTEN    1.597743 -0.169210  0.310849  \n",
       "        ZEUS    2.168410  0.130236  0.178060  \n",
       "        ZUMZ   -1.378972  0.072304  0.157737  \n",
       "\n",
       "[430757 rows x 9 columns]"
      ]
     },
     "execution_count": 102,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.merge(df, sds.swaplevel(), left_index=True, right_index=True)\n",
    "df"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Lag volatility "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>ret</th>\n",
       "      <th>roeq</th>\n",
       "      <th>bm</th>\n",
       "      <th>mom12m</th>\n",
       "      <th>mve</th>\n",
       "      <th>size_rnk</th>\n",
       "      <th>target</th>\n",
       "      <th>predict</th>\n",
       "      <th>sd</th>\n",
       "      <th>lag_sd</th>\n",
       "      <th>inverse_var</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>date</th>\n",
       "      <th>ticker</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2019-05</th>\n",
       "      <th>YELP</th>\n",
       "      <td>-0.232901</td>\n",
       "      <td>-0.092168</td>\n",
       "      <td>-0.396406</td>\n",
       "      <td>-0.581662</td>\n",
       "      <td>15.005158</td>\n",
       "      <td>823.0</td>\n",
       "      <td>-1.145476</td>\n",
       "      <td>0.008860</td>\n",
       "      <td>0.147527</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-06</th>\n",
       "      <th>YELP</th>\n",
       "      <td>0.112268</td>\n",
       "      <td>0.442784</td>\n",
       "      <td>-0.395686</td>\n",
       "      <td>0.082686</td>\n",
       "      <td>14.680300</td>\n",
       "      <td>929.0</td>\n",
       "      <td>0.642551</td>\n",
       "      <td>0.052772</td>\n",
       "      <td>0.147019</td>\n",
       "      <td>0.147527</td>\n",
       "      <td>45.946755</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-07</th>\n",
       "      <th>YELP</th>\n",
       "      <td>0.025453</td>\n",
       "      <td>0.441471</td>\n",
       "      <td>-0.400773</td>\n",
       "      <td>-0.179907</td>\n",
       "      <td>14.786702</td>\n",
       "      <td>918.0</td>\n",
       "      <td>0.356582</td>\n",
       "      <td>0.046187</td>\n",
       "      <td>0.145845</td>\n",
       "      <td>0.147019</td>\n",
       "      <td>46.264752</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-08</th>\n",
       "      <th>YELP</th>\n",
       "      <td>-0.043937</td>\n",
       "      <td>0.447650</td>\n",
       "      <td>-0.402128</td>\n",
       "      <td>0.201198</td>\n",
       "      <td>14.811837</td>\n",
       "      <td>908.0</td>\n",
       "      <td>0.121159</td>\n",
       "      <td>0.054795</td>\n",
       "      <td>0.144772</td>\n",
       "      <td>0.145845</td>\n",
       "      <td>47.012927</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-09</th>\n",
       "      <th>YELP</th>\n",
       "      <td>0.037004</td>\n",
       "      <td>-0.260449</td>\n",
       "      <td>-0.407563</td>\n",
       "      <td>-0.404116</td>\n",
       "      <td>14.682673</td>\n",
       "      <td>932.0</td>\n",
       "      <td>0.046494</td>\n",
       "      <td>0.007121</td>\n",
       "      <td>0.143694</td>\n",
       "      <td>0.144772</td>\n",
       "      <td>47.712104</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     ret      roeq        bm    mom12m        mve  size_rnk  \\\n",
       "date    ticker                                                                \n",
       "2019-05 YELP   -0.232901 -0.092168 -0.396406 -0.581662  15.005158     823.0   \n",
       "2019-06 YELP    0.112268  0.442784 -0.395686  0.082686  14.680300     929.0   \n",
       "2019-07 YELP    0.025453  0.441471 -0.400773 -0.179907  14.786702     918.0   \n",
       "2019-08 YELP   -0.043937  0.447650 -0.402128  0.201198  14.811837     908.0   \n",
       "2019-09 YELP    0.037004 -0.260449 -0.407563 -0.404116  14.682673     932.0   \n",
       "\n",
       "                  target   predict        sd    lag_sd  inverse_var  \n",
       "date    ticker                                                       \n",
       "2019-05 YELP   -1.145476  0.008860  0.147527       NaN          NaN  \n",
       "2019-06 YELP    0.642551  0.052772  0.147019  0.147527    45.946755  \n",
       "2019-07 YELP    0.356582  0.046187  0.145845  0.147019    46.264752  \n",
       "2019-08 YELP    0.121159  0.054795  0.144772  0.145845    47.012927  \n",
       "2019-09 YELP    0.046494  0.007121  0.143694  0.144772    47.712104  "
      ]
     },
     "execution_count": 114,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['lag_sd'] = df.groupby(level=1).sd.shift(1)\n",
    "df.xs('YELP',level='ticker', drop_level=False).head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>ret</th>\n",
       "      <th>roeq</th>\n",
       "      <th>bm</th>\n",
       "      <th>mom12m</th>\n",
       "      <th>mve</th>\n",
       "      <th>size_rnk</th>\n",
       "      <th>target</th>\n",
       "      <th>predict</th>\n",
       "      <th>sd</th>\n",
       "      <th>lag_sd</th>\n",
       "      <th>inverse_var</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>date</th>\n",
       "      <th>ticker</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2019-05</th>\n",
       "      <th>YELP</th>\n",
       "      <td>-0.232901</td>\n",
       "      <td>-0.092168</td>\n",
       "      <td>-0.396406</td>\n",
       "      <td>-0.581662</td>\n",
       "      <td>15.005158</td>\n",
       "      <td>823.0</td>\n",
       "      <td>-1.145476</td>\n",
       "      <td>0.008860</td>\n",
       "      <td>0.147527</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-06</th>\n",
       "      <th>YELP</th>\n",
       "      <td>0.112268</td>\n",
       "      <td>0.442784</td>\n",
       "      <td>-0.395686</td>\n",
       "      <td>0.082686</td>\n",
       "      <td>14.680300</td>\n",
       "      <td>929.0</td>\n",
       "      <td>0.642551</td>\n",
       "      <td>0.052772</td>\n",
       "      <td>0.147019</td>\n",
       "      <td>0.147527</td>\n",
       "      <td>45.946755</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-07</th>\n",
       "      <th>YELP</th>\n",
       "      <td>0.025453</td>\n",
       "      <td>0.441471</td>\n",
       "      <td>-0.400773</td>\n",
       "      <td>-0.179907</td>\n",
       "      <td>14.786702</td>\n",
       "      <td>918.0</td>\n",
       "      <td>0.356582</td>\n",
       "      <td>0.046187</td>\n",
       "      <td>0.145845</td>\n",
       "      <td>0.147019</td>\n",
       "      <td>46.264752</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-08</th>\n",
       "      <th>YELP</th>\n",
       "      <td>-0.043937</td>\n",
       "      <td>0.447650</td>\n",
       "      <td>-0.402128</td>\n",
       "      <td>0.201198</td>\n",
       "      <td>14.811837</td>\n",
       "      <td>908.0</td>\n",
       "      <td>0.121159</td>\n",
       "      <td>0.054795</td>\n",
       "      <td>0.144772</td>\n",
       "      <td>0.145845</td>\n",
       "      <td>47.012927</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-09</th>\n",
       "      <th>YELP</th>\n",
       "      <td>0.037004</td>\n",
       "      <td>-0.260449</td>\n",
       "      <td>-0.407563</td>\n",
       "      <td>-0.404116</td>\n",
       "      <td>14.682673</td>\n",
       "      <td>932.0</td>\n",
       "      <td>0.046494</td>\n",
       "      <td>0.007121</td>\n",
       "      <td>0.143694</td>\n",
       "      <td>0.144772</td>\n",
       "      <td>47.712104</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     ret      roeq        bm    mom12m        mve  size_rnk  \\\n",
       "date    ticker                                                                \n",
       "2019-05 YELP   -0.232901 -0.092168 -0.396406 -0.581662  15.005158     823.0   \n",
       "2019-06 YELP    0.112268  0.442784 -0.395686  0.082686  14.680300     929.0   \n",
       "2019-07 YELP    0.025453  0.441471 -0.400773 -0.179907  14.786702     918.0   \n",
       "2019-08 YELP   -0.043937  0.447650 -0.402128  0.201198  14.811837     908.0   \n",
       "2019-09 YELP    0.037004 -0.260449 -0.407563 -0.404116  14.682673     932.0   \n",
       "\n",
       "                  target   predict        sd    lag_sd  inverse_var  \n",
       "date    ticker                                                       \n",
       "2019-05 YELP   -1.145476  0.008860  0.147527       NaN          NaN  \n",
       "2019-06 YELP    0.642551  0.052772  0.147019  0.147527    45.946755  \n",
       "2019-07 YELP    0.356582  0.046187  0.145845  0.147019    46.264752  \n",
       "2019-08 YELP    0.121159  0.054795  0.144772  0.145845    47.012927  \n",
       "2019-09 YELP    0.046494  0.007121  0.143694  0.144772    47.712104  "
      ]
     },
     "execution_count": 113,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['inverse_var'] = 1/(df.lag_sd**2)\n",
    "df.xs('YELP',level='ticker', drop_level=False).head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Compute returns of portfolios of best and worst stocks"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 146,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df.dropna(subset=[\"predict\"])\n",
    "\n",
    "numstocks = 100\n",
    "\n",
    "df[\"rnk\"] = df.groupby(\"date\").predict.rank(method=\"first\", ascending=False)\n",
    "best = df[df.rnk<=numstocks]\n",
    "\n",
    "df[\"rnk\"] = df.groupby(\"date\").predict.rank(method=\"first\")\n",
    "worst = df[df.rnk<=numstocks]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>ret</th>\n",
       "      <th>roeq</th>\n",
       "      <th>bm</th>\n",
       "      <th>mom12m</th>\n",
       "      <th>mve</th>\n",
       "      <th>size_rnk</th>\n",
       "      <th>target</th>\n",
       "      <th>predict</th>\n",
       "      <th>sd</th>\n",
       "      <th>lag_sd</th>\n",
       "      <th>inverse_var</th>\n",
       "      <th>rnk</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>date</th>\n",
       "      <th>ticker</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"5\" valign=\"top\">2005-01</th>\n",
       "      <th>AAII</th>\n",
       "      <td>-0.178462</td>\n",
       "      <td>2.118992</td>\n",
       "      <td>-1.437618</td>\n",
       "      <td>-3.087117</td>\n",
       "      <td>11.439327</td>\n",
       "      <td>3186.0</td>\n",
       "      <td>-1.292443</td>\n",
       "      <td>-0.176312</td>\n",
       "      <td>0.244647</td>\n",
       "      <td>0.245369</td>\n",
       "      <td>16.609714</td>\n",
       "      <td>83.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ABFI</th>\n",
       "      <td>-0.466667</td>\n",
       "      <td>-2.880181</td>\n",
       "      <td>0.766901</td>\n",
       "      <td>-0.567059</td>\n",
       "      <td>8.370455</td>\n",
       "      <td>4236.0</td>\n",
       "      <td>-2.840673</td>\n",
       "      <td>-0.273414</td>\n",
       "      <td>0.249967</td>\n",
       "      <td>0.244860</td>\n",
       "      <td>16.678775</td>\n",
       "      <td>12.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ACEC</th>\n",
       "      <td>0.291667</td>\n",
       "      <td>-1.874422</td>\n",
       "      <td>-1.156635</td>\n",
       "      <td>-0.574120</td>\n",
       "      <td>10.407385</td>\n",
       "      <td>3818.0</td>\n",
       "      <td>2.191251</td>\n",
       "      <td>-0.169931</td>\n",
       "      <td>0.347926</td>\n",
       "      <td>0.349213</td>\n",
       "      <td>8.200097</td>\n",
       "      <td>91.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ACR</th>\n",
       "      <td>0.064461</td>\n",
       "      <td>-2.059313</td>\n",
       "      <td>-1.859203</td>\n",
       "      <td>2.011144</td>\n",
       "      <td>12.608771</td>\n",
       "      <td>2237.0</td>\n",
       "      <td>1.112948</td>\n",
       "      <td>-0.267857</td>\n",
       "      <td>0.213641</td>\n",
       "      <td>0.215392</td>\n",
       "      <td>21.554619</td>\n",
       "      <td>16.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AICX</th>\n",
       "      <td>0.090909</td>\n",
       "      <td>-2.127976</td>\n",
       "      <td>-1.084324</td>\n",
       "      <td>-2.235167</td>\n",
       "      <td>9.441300</td>\n",
       "      <td>4132.0</td>\n",
       "      <td>1.321948</td>\n",
       "      <td>-0.276635</td>\n",
       "      <td>0.365792</td>\n",
       "      <td>0.368810</td>\n",
       "      <td>7.351819</td>\n",
       "      <td>11.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     ret      roeq        bm    mom12m        mve  size_rnk  \\\n",
       "date    ticker                                                                \n",
       "2005-01 AAII   -0.178462  2.118992 -1.437618 -3.087117  11.439327    3186.0   \n",
       "        ABFI   -0.466667 -2.880181  0.766901 -0.567059   8.370455    4236.0   \n",
       "        ACEC    0.291667 -1.874422 -1.156635 -0.574120  10.407385    3818.0   \n",
       "        ACR     0.064461 -2.059313 -1.859203  2.011144  12.608771    2237.0   \n",
       "        AICX    0.090909 -2.127976 -1.084324 -2.235167   9.441300    4132.0   \n",
       "\n",
       "                  target   predict        sd    lag_sd  inverse_var   rnk  \n",
       "date    ticker                                                             \n",
       "2005-01 AAII   -1.292443 -0.176312  0.244647  0.245369    16.609714  83.0  \n",
       "        ABFI   -2.840673 -0.273414  0.249967  0.244860    16.678775  12.0  \n",
       "        ACEC    2.191251 -0.169931  0.347926  0.349213     8.200097  91.0  \n",
       "        ACR     1.112948 -0.267857  0.213641  0.215392    21.554619  16.0  \n",
       "        AICX    1.321948 -0.276635  0.365792  0.368810     7.351819  11.0  "
      ]
     },
     "execution_count": 132,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "worst.head()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Equal-weighted returns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 133,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>best</th>\n",
       "      <th>worst</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>date</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2005-01</th>\n",
       "      <td>-0.005356</td>\n",
       "      <td>-0.081827</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2005-02</th>\n",
       "      <td>0.041530</td>\n",
       "      <td>-0.074315</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2005-03</th>\n",
       "      <td>-0.048466</td>\n",
       "      <td>-0.089032</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2005-04</th>\n",
       "      <td>-0.061057</td>\n",
       "      <td>-0.105959</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2005-05</th>\n",
       "      <td>0.051723</td>\n",
       "      <td>0.043296</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             best     worst\n",
       "date                       \n",
       "2005-01 -0.005356 -0.081827\n",
       "2005-02  0.041530 -0.074315\n",
       "2005-03 -0.048466 -0.089032\n",
       "2005-04 -0.061057 -0.105959\n",
       "2005-05  0.051723  0.043296"
      ]
     },
     "execution_count": 133,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "best_rets  = best.groupby(\"date\").ret.mean()\n",
    "worst_rets = worst.groupby(\"date\").ret.mean()\n",
    "rets = pd.concat((best_rets, worst_rets), axis=1)\n",
    "rets.columns = [\"best\", \"worst\"]\n",
    "rets.head()\n",
    "# rets.to_csv(\"files/rets.csv\")\n"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Risk-parity returns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 147,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>ret</th>\n",
       "      <th>roeq</th>\n",
       "      <th>bm</th>\n",
       "      <th>mom12m</th>\n",
       "      <th>mve</th>\n",
       "      <th>size_rnk</th>\n",
       "      <th>target</th>\n",
       "      <th>predict</th>\n",
       "      <th>sd</th>\n",
       "      <th>lag_sd</th>\n",
       "      <th>inverse_var</th>\n",
       "      <th>rnk</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>date</th>\n",
       "      <th>ticker</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"5\" valign=\"top\">2005-01</th>\n",
       "      <th>ABL</th>\n",
       "      <td>0.052632</td>\n",
       "      <td>1.182907</td>\n",
       "      <td>1.628030</td>\n",
       "      <td>1.130569</td>\n",
       "      <td>10.577421</td>\n",
       "      <td>3731.0</td>\n",
       "      <td>1.008471</td>\n",
       "      <td>0.170241</td>\n",
       "      <td>0.096840</td>\n",
       "      <td>0.097448</td>\n",
       "      <td>105.306743</td>\n",
       "      <td>6.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ACAS</th>\n",
       "      <td>0.019490</td>\n",
       "      <td>1.338645</td>\n",
       "      <td>0.419595</td>\n",
       "      <td>0.286749</td>\n",
       "      <td>14.900129</td>\n",
       "      <td>626.0</td>\n",
       "      <td>0.695581</td>\n",
       "      <td>0.116341</td>\n",
       "      <td>0.079592</td>\n",
       "      <td>0.080264</td>\n",
       "      <td>155.225159</td>\n",
       "      <td>93.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AGII</th>\n",
       "      <td>0.013725</td>\n",
       "      <td>0.389729</td>\n",
       "      <td>1.641080</td>\n",
       "      <td>0.577761</td>\n",
       "      <td>13.280026</td>\n",
       "      <td>1721.0</td>\n",
       "      <td>0.631787</td>\n",
       "      <td>0.127166</td>\n",
       "      <td>0.104596</td>\n",
       "      <td>0.105478</td>\n",
       "      <td>89.882681</td>\n",
       "      <td>63.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ALDA</th>\n",
       "      <td>-0.072787</td>\n",
       "      <td>1.483048</td>\n",
       "      <td>1.812573</td>\n",
       "      <td>2.518106</td>\n",
       "      <td>11.266855</td>\n",
       "      <td>3301.0</td>\n",
       "      <td>-0.478855</td>\n",
       "      <td>0.150051</td>\n",
       "      <td>0.193612</td>\n",
       "      <td>0.194712</td>\n",
       "      <td>26.376371</td>\n",
       "      <td>25.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ALG</th>\n",
       "      <td>-0.033137</td>\n",
       "      <td>0.426933</td>\n",
       "      <td>1.259061</td>\n",
       "      <td>0.823644</td>\n",
       "      <td>12.485536</td>\n",
       "      <td>2336.0</td>\n",
       "      <td>-0.015578</td>\n",
       "      <td>0.116249</td>\n",
       "      <td>0.069438</td>\n",
       "      <td>0.069686</td>\n",
       "      <td>205.926709</td>\n",
       "      <td>94.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"5\" valign=\"top\">2022-03</th>\n",
       "      <th>VIA</th>\n",
       "      <td>-0.260987</td>\n",
       "      <td>1.335621</td>\n",
       "      <td>-0.000832</td>\n",
       "      <td>0.692222</td>\n",
       "      <td>12.069410</td>\n",
       "      <td>1858.0</td>\n",
       "      <td>-2.049083</td>\n",
       "      <td>0.087820</td>\n",
       "      <td>0.093421</td>\n",
       "      <td>0.090719</td>\n",
       "      <td>121.508051</td>\n",
       "      <td>100.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>VIRC</th>\n",
       "      <td>-0.006645</td>\n",
       "      <td>1.091610</td>\n",
       "      <td>1.286850</td>\n",
       "      <td>-0.051664</td>\n",
       "      <td>10.788639</td>\n",
       "      <td>2191.0</td>\n",
       "      <td>-0.024321</td>\n",
       "      <td>0.103784</td>\n",
       "      <td>0.128988</td>\n",
       "      <td>0.129351</td>\n",
       "      <td>59.767118</td>\n",
       "      <td>18.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>WSBF</th>\n",
       "      <td>-0.010235</td>\n",
       "      <td>0.660527</td>\n",
       "      <td>0.661027</td>\n",
       "      <td>0.535450</td>\n",
       "      <td>13.067852</td>\n",
       "      <td>1602.0</td>\n",
       "      <td>-0.063724</td>\n",
       "      <td>0.090299</td>\n",
       "      <td>0.130029</td>\n",
       "      <td>0.130415</td>\n",
       "      <td>58.795838</td>\n",
       "      <td>75.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>X</th>\n",
       "      <td>0.386990</td>\n",
       "      <td>1.981783</td>\n",
       "      <td>0.925956</td>\n",
       "      <td>0.871628</td>\n",
       "      <td>15.775596</td>\n",
       "      <td>586.0</td>\n",
       "      <td>1.985105</td>\n",
       "      <td>0.109824</td>\n",
       "      <td>0.177038</td>\n",
       "      <td>0.175901</td>\n",
       "      <td>32.319572</td>\n",
       "      <td>12.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ZEUS</th>\n",
       "      <td>0.436683</td>\n",
       "      <td>1.467827</td>\n",
       "      <td>1.924874</td>\n",
       "      <td>0.666564</td>\n",
       "      <td>12.604142</td>\n",
       "      <td>1728.0</td>\n",
       "      <td>2.168410</td>\n",
       "      <td>0.130236</td>\n",
       "      <td>0.178060</td>\n",
       "      <td>0.176567</td>\n",
       "      <td>32.076151</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>20700 rows × 12 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                     ret      roeq        bm    mom12m        mve  size_rnk  \\\n",
       "date    ticker                                                                \n",
       "2005-01 ABL     0.052632  1.182907  1.628030  1.130569  10.577421    3731.0   \n",
       "        ACAS    0.019490  1.338645  0.419595  0.286749  14.900129     626.0   \n",
       "        AGII    0.013725  0.389729  1.641080  0.577761  13.280026    1721.0   \n",
       "        ALDA   -0.072787  1.483048  1.812573  2.518106  11.266855    3301.0   \n",
       "        ALG    -0.033137  0.426933  1.259061  0.823644  12.485536    2336.0   \n",
       "...                  ...       ...       ...       ...        ...       ...   \n",
       "2022-03 VIA    -0.260987  1.335621 -0.000832  0.692222  12.069410    1858.0   \n",
       "        VIRC   -0.006645  1.091610  1.286850 -0.051664  10.788639    2191.0   \n",
       "        WSBF   -0.010235  0.660527  0.661027  0.535450  13.067852    1602.0   \n",
       "        X       0.386990  1.981783  0.925956  0.871628  15.775596     586.0   \n",
       "        ZEUS    0.436683  1.467827  1.924874  0.666564  12.604142    1728.0   \n",
       "\n",
       "                  target   predict        sd    lag_sd  inverse_var    rnk  \n",
       "date    ticker                                                              \n",
       "2005-01 ABL     1.008471  0.170241  0.096840  0.097448   105.306743    6.0  \n",
       "        ACAS    0.695581  0.116341  0.079592  0.080264   155.225159   93.0  \n",
       "        AGII    0.631787  0.127166  0.104596  0.105478    89.882681   63.0  \n",
       "        ALDA   -0.478855  0.150051  0.193612  0.194712    26.376371   25.0  \n",
       "        ALG    -0.015578  0.116249  0.069438  0.069686   205.926709   94.0  \n",
       "...                  ...       ...       ...       ...          ...    ...  \n",
       "2022-03 VIA    -2.049083  0.087820  0.093421  0.090719   121.508051  100.0  \n",
       "        VIRC   -0.024321  0.103784  0.128988  0.129351    59.767118   18.0  \n",
       "        WSBF   -0.063724  0.090299  0.130029  0.130415    58.795838   75.0  \n",
       "        X       1.985105  0.109824  0.177038  0.175901    32.319572   12.0  \n",
       "        ZEUS    2.168410  0.130236  0.178060  0.176567    32.076151    3.0  \n",
       "\n",
       "[20700 rows x 12 columns]"
      ]
     },
     "execution_count": 147,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "best"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 148,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "date\n",
       "2005-01    5611.806555\n",
       "2005-02    6029.050128\n",
       "2005-03    5454.688393\n",
       "2005-04    5507.294438\n",
       "2005-05    5981.954414\n",
       "              ...     \n",
       "2021-11    9222.628044\n",
       "2021-12    7705.782159\n",
       "2022-01    7348.069793\n",
       "2022-02    8236.535117\n",
       "2022-03    7600.272952\n",
       "Name: sum_inverse_var, Length: 207, dtype: float64"
      ]
     },
     "execution_count": 148,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sum_inverse_var = best.groupby('date').inverse_var.sum()\n",
    "sum_inverse_var.name = 'sum_inverse_var'\n",
    "sum_inverse_var"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 149,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>date</th>\n",
       "      <th>ticker</th>\n",
       "      <th>ret</th>\n",
       "      <th>roeq</th>\n",
       "      <th>bm</th>\n",
       "      <th>mom12m</th>\n",
       "      <th>mve</th>\n",
       "      <th>size_rnk</th>\n",
       "      <th>target</th>\n",
       "      <th>predict</th>\n",
       "      <th>sd</th>\n",
       "      <th>lag_sd</th>\n",
       "      <th>inverse_var</th>\n",
       "      <th>rnk</th>\n",
       "      <th>sum_inverse_var</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2005-01</td>\n",
       "      <td>ABL</td>\n",
       "      <td>0.052632</td>\n",
       "      <td>1.182907</td>\n",
       "      <td>1.628030</td>\n",
       "      <td>1.130569</td>\n",
       "      <td>10.577421</td>\n",
       "      <td>3731.0</td>\n",
       "      <td>1.008471</td>\n",
       "      <td>0.170241</td>\n",
       "      <td>0.096840</td>\n",
       "      <td>0.097448</td>\n",
       "      <td>105.306743</td>\n",
       "      <td>6.0</td>\n",
       "      <td>5611.806555</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2005-01</td>\n",
       "      <td>ACAS</td>\n",
       "      <td>0.019490</td>\n",
       "      <td>1.338645</td>\n",
       "      <td>0.419595</td>\n",
       "      <td>0.286749</td>\n",
       "      <td>14.900129</td>\n",
       "      <td>626.0</td>\n",
       "      <td>0.695581</td>\n",
       "      <td>0.116341</td>\n",
       "      <td>0.079592</td>\n",
       "      <td>0.080264</td>\n",
       "      <td>155.225159</td>\n",
       "      <td>93.0</td>\n",
       "      <td>5611.806555</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2005-01</td>\n",
       "      <td>AGII</td>\n",
       "      <td>0.013725</td>\n",
       "      <td>0.389729</td>\n",
       "      <td>1.641080</td>\n",
       "      <td>0.577761</td>\n",
       "      <td>13.280026</td>\n",
       "      <td>1721.0</td>\n",
       "      <td>0.631787</td>\n",
       "      <td>0.127166</td>\n",
       "      <td>0.104596</td>\n",
       "      <td>0.105478</td>\n",
       "      <td>89.882681</td>\n",
       "      <td>63.0</td>\n",
       "      <td>5611.806555</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2005-01</td>\n",
       "      <td>ALDA</td>\n",
       "      <td>-0.072787</td>\n",
       "      <td>1.483048</td>\n",
       "      <td>1.812573</td>\n",
       "      <td>2.518106</td>\n",
       "      <td>11.266855</td>\n",
       "      <td>3301.0</td>\n",
       "      <td>-0.478855</td>\n",
       "      <td>0.150051</td>\n",
       "      <td>0.193612</td>\n",
       "      <td>0.194712</td>\n",
       "      <td>26.376371</td>\n",
       "      <td>25.0</td>\n",
       "      <td>5611.806555</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2005-01</td>\n",
       "      <td>ALG</td>\n",
       "      <td>-0.033137</td>\n",
       "      <td>0.426933</td>\n",
       "      <td>1.259061</td>\n",
       "      <td>0.823644</td>\n",
       "      <td>12.485536</td>\n",
       "      <td>2336.0</td>\n",
       "      <td>-0.015578</td>\n",
       "      <td>0.116249</td>\n",
       "      <td>0.069438</td>\n",
       "      <td>0.069686</td>\n",
       "      <td>205.926709</td>\n",
       "      <td>94.0</td>\n",
       "      <td>5611.806555</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      date ticker       ret      roeq        bm    mom12m        mve  \\\n",
       "0  2005-01    ABL  0.052632  1.182907  1.628030  1.130569  10.577421   \n",
       "1  2005-01   ACAS  0.019490  1.338645  0.419595  0.286749  14.900129   \n",
       "2  2005-01   AGII  0.013725  0.389729  1.641080  0.577761  13.280026   \n",
       "3  2005-01   ALDA -0.072787  1.483048  1.812573  2.518106  11.266855   \n",
       "4  2005-01    ALG -0.033137  0.426933  1.259061  0.823644  12.485536   \n",
       "\n",
       "   size_rnk    target   predict        sd    lag_sd  inverse_var   rnk  \\\n",
       "0    3731.0  1.008471  0.170241  0.096840  0.097448   105.306743   6.0   \n",
       "1     626.0  0.695581  0.116341  0.079592  0.080264   155.225159  93.0   \n",
       "2    1721.0  0.631787  0.127166  0.104596  0.105478    89.882681  63.0   \n",
       "3    3301.0 -0.478855  0.150051  0.193612  0.194712    26.376371  25.0   \n",
       "4    2336.0 -0.015578  0.116249  0.069438  0.069686   205.926709  94.0   \n",
       "\n",
       "   sum_inverse_var  \n",
       "0      5611.806555  \n",
       "1      5611.806555  \n",
       "2      5611.806555  \n",
       "3      5611.806555  \n",
       "4      5611.806555  "
      ]
     },
     "execution_count": 149,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "best = best.reset_index().merge(sum_inverse_var,left_on='date',right_on='date')\n",
    "best.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 153,
   "metadata": {},
   "outputs": [],
   "source": [
    "best['wgt'] = best.inverse_var/best.sum_inverse_var\n",
    "best['wgt_x_ret'] = best.wgt * best.ret\n",
    "best_rets = best.groupby(\"date\").wgt_x_ret.sum()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Same for worst"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 154,
   "metadata": {},
   "outputs": [],
   "source": [
    "sum_inverse_var = worst.groupby('date').inverse_var.sum()\n",
    "sum_inverse_var.name = 'sum_inverse_var'\n",
    "worst = worst.reset_index().merge(sum_inverse_var,left_on='date',right_on='date')\n",
    "worst['wgt'] = worst.inverse_var/worst.sum_inverse_var\n",
    "worst['wgt_x_ret'] = worst.wgt * worst.ret\n",
    "worst_rets = worst.groupby(\"date\").wgt_x_ret.sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 156,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>best</th>\n",
       "      <th>worst</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>date</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2005-01</th>\n",
       "      <td>0.009367</td>\n",
       "      <td>-0.069581</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2005-02</th>\n",
       "      <td>0.045342</td>\n",
       "      <td>-0.051452</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2005-03</th>\n",
       "      <td>-0.043387</td>\n",
       "      <td>-0.062486</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2005-04</th>\n",
       "      <td>-0.044839</td>\n",
       "      <td>-0.071012</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2005-05</th>\n",
       "      <td>0.053237</td>\n",
       "      <td>0.039750</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             best     worst\n",
       "date                       \n",
       "2005-01  0.009367 -0.069581\n",
       "2005-02  0.045342 -0.051452\n",
       "2005-03 -0.043387 -0.062486\n",
       "2005-04 -0.044839 -0.071012\n",
       "2005-05  0.053237  0.039750"
      ]
     },
     "execution_count": 156,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rets_risk_parity = pd.concat((best_rets, worst_rets), axis=1)\n",
    "rets_risk_parity.columns = [\"best\", \"worst\"]\n",
    "rets_risk_parity.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 162,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Equal-weighted\n",
      "Best: average return:\t  0.0134\n",
      "Best: SD(return):\t  0.0552\n",
      "Worst: average return:\t  0.0134\n",
      "Worst: SD(return):\t  0.1142\n",
      "\n",
      "Risk parity\n",
      "Best: average return:\t  0.0118\n",
      "Best: SD(return):\t  0.0453\n",
      "Worst: average return:\t  0.0144\n",
      "Worst: SD(return):\t  0.1117\n"
     ]
    }
   ],
   "source": [
    "print(f'Equal-weighted')\n",
    "print(f'Best: average return:\\t {rets.best.mean(): .4f}')\n",
    "print(f'Best: SD(return):\\t {rets.best.std(): .4f}')\n",
    "print(f'Worst: average return:\\t {rets.worst.mean(): .4f}')\n",
    "print(f'Worst: SD(return):\\t {rets.worst.std(): .4f}')\n",
    "\n",
    "print(f'\\nRisk parity')\n",
    "print(f'Best: average return:\\t {rets_risk_parity.best.mean(): .4f}')\n",
    "print(f'Best: SD(return):\\t {rets_risk_parity.best.std(): .4f}')\n",
    "print(f'Worst: average return:\\t {rets_risk_parity.worst.mean(): .4f}')\n",
    "print(f'Worst: SD(return):\\t {rets_risk_parity.worst.std(): .4f}')\n",
    "\n"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Performance evaluation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 171,
   "metadata": {},
   "outputs": [],
   "source": [
    "from pandas_datareader import DataReader as pdr\n",
    "ff = pdr(\"F-F_Research_Data_5_Factors_2x3\", \"famafrench\", start=2005)[0]/100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 177,
   "metadata": {},
   "outputs": [],
   "source": [
    "rets.index = pd.to_datetime(rets.index).to_period(\"M\")\n",
    "rets[\"mkt\"] = ff[\"Mkt-RF\"] + ff[\"RF\"]\n",
    "rets[\"ls\"] = 1.3*rets[\"best\"] - 0.3*rets[\"worst\"]\n",
    "rets[\"rf\"] = ff[\"RF\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 185,
   "metadata": {},
   "outputs": [],
   "source": [
    "rets_risk_parity.index = pd.to_datetime(rets_risk_parity.index).to_period(\"M\")\n",
    "rets_risk_parity[\"mkt\"] = ff[\"Mkt-RF\"] + ff[\"RF\"]\n",
    "rets_risk_parity[\"ls\"] = 1.3*rets_risk_parity[\"best\"] - 0.3*rets_risk_parity[\"worst\"]\n",
    "rets_risk_parity[\"rf\"] = ff[\"RF\"]"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Accumulations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 184,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='date'>"
      ]
     },
     "execution_count": 184,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "(1+rets[[\"best\", \"worst\", \"mkt\", \"ls\"]]).cumprod().plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 186,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='date'>"
      ]
     },
     "execution_count": 186,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWoAAAEGCAYAAABM7t/CAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8qNh9FAAAACXBIWXMAAAsTAAALEwEAmpwYAABv1ElEQVR4nO2ddZiUVfvHP2dqu5tNWLq7OwUMVFQU9Wd3d+tr+2K9KgaKgigWCIiAhHR3N2x31+zs1PP749ldWGB7ll3Y87kurp154jz3M8N858x97hCKoiCRSCSSpoumsQ2QSCQSSdVIoZZIJJImjhRqiUQiaeJIoZZIJJImjhRqiUQiaeLoGmJQf39/JSoqqiGGlkgkksuSXbt2ZSqKEnChfQ0i1FFRUezcubMhhpZIJJLLEiFEXGX7pOtDIpFImjhSqCUSiaSJI4VaIpFImjgN4qO+EBaLhcTEREwm08W6ZJPA2dmZsLAw9Hp9Y5sikUguUS6aUCcmJuLh4UFUVBRCiIt12UZFURSysrJITEykZcuWjW2ORCK5RLlorg+TyYSfn1+zEWkAIQR+fn7N7leERCJxLBfVR92cRLqM5njPEonEscjFRIlEIrmIWGx25m6Lx2y11/icZiPUsbGxdO7cuV5jrF27ls2bNzvIIolE0hxZczSdlxYc4O/9yTU+p9kItSOQQi2RSOrLkZQCAFYdSavxOc1KqK1WK1OnTqVDhw5MnjwZo9HIrl27GDZsGL169WLcuHGkpKQA8Nlnn9GxY0e6du3KlClTiI2N5euvv+aTTz6he/fubNiwoZHvRiKRXIocSckHYN2xDEqsthqdc9HC887mP4sPcTg536FjdmzhyetXdarymGPHjjFz5kwGDRrEXXfdxfTp01mwYAGLFi0iICCA3377jZdffpnvv/+e999/n5iYGJycnMjNzcXb25sHHngAd3d3nnnmGYfaLpFImg9HUvPxdTOQXWRm86ksRrQLrPacZjWjDg8PZ9CgQQDceuutLF++nIMHDzJmzBi6d+/O22+/TWJiIgBdu3Zl6tSp/PTTT+h0jfJ9JpFILjOKSqzEZRm5pW8ErgYtqw7XzP3RKApU3cy3oTg3VM7Dw4NOnTqxZcuW845dsmQJ69evZ/HixbzzzjscOHDgYpkpkUguM7bHZLNgTxLX9wwFoFu4NyPaBfLXvmTuHdKKKH+3Ks9vVjPq+Pj4clGeO3cu/fv3JyMjo3ybxWLh0KFD2O12EhISGDFiBB988AF5eXkUFhbi4eFBQUFBY96CRCK5BFmwJ4lftsczbfkxANoHe/D8Fe3RagT3/LiTB+bsqvL8ZiXU7dq1Y/r06XTo0IGcnBweffRR5s2bx/PPP0+3bt3o3r07mzdvxmazceutt9KlSxd69OjBY489hre3N1dddRULFiyQi4kSiaRWnExXJ3jbYrLxcNIR5uNChJ8rX07tSUxmEZtOZVZ5vlAUxeFG9e7dWzm3ccCRI0fo0KGDw691KdCc710iae4oikL3N1fi6aIjIbuYPlE+/PHAwPL9MZlF+Lsb8HQx7FIUpfeFxpCrZBKJRNKAZBaaySu28NioNmw7ncWAaL8K+1tW458GKdQSiUTSoJwodXu0DXLn7sF1q6LZrHzUEolEcrE5lV4IQJtAjzqPIYVaIpFIGpAT6YV4OOkI8nSq8xhSqCUSiaQBOZFWSOsg93qVPJZCLZFILhqKolBy8mRjm3FROZFeSJtA93qNIYXaASxcuJDDhw83thkSSZPHuG07p6+8ioLVqxvblItCRkEJmYUl9fJPgxTqWmGzXbjSlRRqiaRmGHerGXjZs2Y3siUXh5kbYxAChrULqNc4zUaop02bxmeffQbAk08+yciRIwFYvXo1U6dO5ZdffqFLly507tyZ559/vvw8d3d3nn76abp168aWLVt44YUXysufPvPMM2zevJm//vqLZ599lu7du3Pq1KlGuT+J5FLAdPAQAMbt2zEdO9bI1jQsafkmftgUw6TuobQNqt+MunHiqJe9AKkOLnIU3AXGv1/p7iFDhvDRRx/x2GOPsXPnTkpKSrBYLGzYsIG2bdvy/PPPs2vXLnx8fBg7diwLFy5k0qRJFBUV0a9fPz766COysrK4++67OXr0KEKI8vKnV199NVdeeSWTJ0927D1JJJcZpoMHcRs2FOO27WTPmUOLt99ubJMajOlrTmJXFJ4c3bbeYzWbGXWvXr3YtWsX+fn5ODk5MWDAAHbu3MmGDRvw9vZm+PDhBAQEoNPpmDp1KuvXrwdAq9Vy/fXXA+Dl5YWzszN33303f/75J66uro15SxLJJYUlLR1rejrugwbhOW4sBStX0RAlLJoCNrvC4n3JTOwSQoRf/XWicWbUVcx8Gwq9Xk/Lli2ZNWsWAwcOpGvXrqxZs4aTJ08SFRXFrl0Xrl7l7OyMVqsFQKfTsX37dv7991/mzZvHF198wepmsigikdQX06GDADh37oJwdiZv0V9Y4uMxREY2smWOZ29CLjlGC6M6BDlkvGYzowbV/fHhhx8ydOhQhgwZwtdff02PHj3o27cv69atIzMzE5vNxi+//MKwYcPOO7+wsJC8vDwmTJjAJ598wr59+wBk+VOJpAaYDh4EjQbnDu1x6doVgOL9l2ed97XH0tEIGNqmfouIZTQ7oU5JSWHAgAEEBQXh7OzMkCFDCAkJ4f3332fEiBF069aNXr16cc0115x3fkFBAVdeeSVdu3Zl8ODBfPzxxwBMmTKFadOm0aNHD7mYKJFUQvHBgzi1bo3GxQWn1q0RLi4UH9hfqzEURSHpqacpbGJlhk+mF7BwT1J5D8Q1x9LpFemDl6veIeM3q6JMo0aNwmKxlD8/fvx4+eObb76Zm2+++bxzCgsLyx+HhISwffv2844ZNGiQDM+TSKrAkpaOcecuvK68EgCh0+HcqSOmWs6obVlZ5C9diunoUdwGDUJomsZc872lR/n3aDof/OPMdT1DOZiUz7Pj2jls/KZxlxKJ5LIm/b//BasVv3vvKd/m0qUrpsOHUc6aPFWHpbSnqfn0aQqbyPqQoijsS8ylT5QPUX5uTF+j/qoe2b7yprWK1Ypx1y6KDx2q0TWa1YxaIpFcfIq2bSd/yRL8H3oIQ3h4+XaXrl3I/sGM6fhxXDrVrI+qOTEJAI2bG5kzvsV91Kh61dBwBEm5xWQWmnl8dFtu6x9JSl4x8VlGOoR4XvB4c0ICsTfehC0nB+HqSut/V6Hz8anyGnJGLZFIGgzFYiH1rTfRh4bid9+9FfY5dyldUCxdlK8JZTNqv3vvxbR/f/nzxmRfQh4A3cK8AAjxcqFfK79Kjy/euw9bTg7+jzyCYjSSM+enaq8hhVoikTQY2T/9jPnkKYJefgmNs3OFffrQFujDwihc9W+141iSkkr/JqL19cWlRw/1+QWE2hwXhzUjwwHW14x9ibkYtBraB194Bn0u5vg4EAK/e+/BY8xosn/6CVs1UWNSqCUSSYNgy8sj84svcBs2FPcRI87bL4TA88qJFG3dWqWwGnfv4eSo0Rj37MGSlIQ+LAx9aCgA5nOE2lZQQMwNN3Jy3BVkzZx5URJq9ibk0rGFJwZdzeTUEp+ALjgYjZMTfvc/gD0/n5y5v1R5To1GFkI8KYQ4JIQ4KIT4RQjhXP1ZEomkOVO0dRv2oiL877+/Uj+y11VXgd1O/rJllY5jKg3hM27dijkxCUNYKPrgINBqy2faZeTM/QV7fj4uHTuSPu1DinfvdtwNXQCrzc6BxDy6h3vX+BxzfHy5r96lcyfchg4he9asKs+pVqiFEKHAY0BvRVE6A1pgSo2tuoR44403+PDDD8/bPmvWLJKTkxvBIonk0sW4bSvC1RWXLl0qPcYpOhqnjh3IW/x3he3m+Hhy/1wAUF6/2rhzF5aUFPShYQidDn1wMJbEM0JtLy4me/Zs3IYMIeT999RxYmMdfFcVOZlRSLHFRrdwL1AUWPQIxG6s8hxzfDyGyIjy5/4PPIgtJ6fKc2rq+tABLkIIHeAKNCvVkkItkdSeom3bce3VC6GvOunD68qrMB04gDkhoXxb9qzZpLz0EtbMTEqOnygdbxtYLOjDwgDQh4VVmFHnLVyILTsb//vvQx8UBBrNeTNuR7M9JhuAXhG+kBsHe+bAscp/HdgKi7BlZaGPOCPUrj174NqvX5XXqVaoFUVJAj4E4oEUIE9RlBXnHieEuE8IsVMIsTPjIjrya0psbCzt27fnjjvuoG3btkydOpVVq1YxaNAg2rRpc14iy7fffsv48eOZM2cOO3fuZOrUqXTv3p3i4uJGugOJpOlTcuoUSc89R8npGMynTuHWr2+153iMUksOF65dV77NdFwtgWrcvZuSkyfRenuD1QqAPkz1T+tDQyssJuYvX4GhdTSuvXsj9Hp0QUENLtSbT2YR6u1CuK8LpJRGrxSmVXq8JSEeAEN4RIXt/g8+UOV1qo2jFkL4ANcALYFc4A8hxK2KolSIKVEUZQYwA6B3795VevA/2P4BR7OPVnfpWtHetz3P932+ymNOnjzJH3/8wffff0+fPn2YO3cuGzdu5K+//uLdd9+le/fuAHzxxResXLmShQsX4uTkxMyZM/nwww/p3bu3Q22WSC43ClasIP+vxRh37gTAtV//as8xREZiiIqicN06fG+7VW3XVTqLzl+yFHtREb533kn2Dz+ox5cuJOrDQrFmZGAvKUExmzHu3InfnXeUj6sPbYG5AYXablfYGpPFmA5Bqg++XKjTKz3HHFcq1JEVhbreM2pgNBCjKEqGoigW4E9gYA3Oa3K0bNmSLl26oNFo6NSpE6NKg+W7dOlCbKkv68cff2TZsmXMmzcPJ6e6dw2WSJojptKyDNbkFDQeHjh37FCj89yHDcO4fTt2oxFrWhr2/HwACv5VQ/c8Ro5A6+cHQqBr0QI4I9iWpGSKNm0CqxX34cPLxzSEhmJpQJflkdR8co0WBkSXxkyXCXVR5R4Fc7wq1PpzZtTVJe3UJDMxHugvhHAFioFRwM4anFcp1c18G4qzhVej0ZQ/12g0WEt/VnXp0oW9e/eSmJhIy5YtG8VOieRSpeTECdyHDQOdDl1gAKK0RHB1uA8fRvbs2RRt3Vp+jkvvXhTvVMsPO7Vpg2vv3pgOH0ZjMACU+6otSUkUrlmL1ssLl27dysfUh4ZiXfw3isVSrZ+8Lmw5lQWgCrWiQPJedUc1rg+tnx9ad7daXataoVYUZZsQYh6wG7ACeyh1cVyO9OjRgwcffJCrr76a5cuX06JFC1nGVCKpAXazGXNsHB6jRhP45BO1Ote1Vy80rq4UrlmLPkINXfO5aQrFO3ehDfBH6+1N0Msvlc+0gTOx1AnxFK5fj9uwoQidruJ+ux1LWhqGUlF3JJtPZdHK340QLxfITwZjJrj6gTEbbFbQni+v5rh4DBERFxitamoU9aEoyuuKorRXFKWzoii3KYpSUusrXUIMHjyYDz/8kIkTJ5KZmckdd9zBAw88IBcTJZIqMMfEgtWKU5s2tT5XGAy4jx5F3pIlGLduQxcSgvvQISAETq1bA6APDCx/DKALDAS9nszpX2LLycFz3LgKY+pLXSRnh/A5kj3xOfSJ8lWflLk9Wo8GFFW0z6Fo61aK9+2rcA81pdkUZYqKiuLgwYPlz2edFWB+7j6AcePGMa70jb/++uvL23FJJJKKpL71NqYjR/ApLRNcF6EG8L//fvIX/03Rpk24DRuK1ssLr+uvw7V0kf9chEaDvkUIlrh4fG65GffShtVl6Mt92I4X6uwiMzlGC22DS5vWpuwDBESPhP2/qe4Pj+Dy44179pBw/wMYIiIIePyxWl+v2Qi1RCJxPKZjx8mZOxcUBcVmBZ0Op5ZRdRrLKToazwkTyF+yBOe2ai3n6prfug8ZirV9BkEvvXTegpw+OBiEaJAFxZPpap366AA3sJhg3y/Qojv4RKkHFFZcUMz/ewlCqyXix9nVVsq7EFKoJRJJnUn/6EM0Hh4IvR7Tvv0YWkcjShf76oL/ww9RsGYNrn1qFgob/MrLle4TBkODxVKfylCFunWgO2z5HHJi4ar/gVtp662iiiF65vh4DC1b1kmkQQq1RCKpA4WbNpH17XcYt24l8NlnsRuNZE6fXme3RxlOrVrRbttWh0Vp6ENDG0SoT6YX4qLX0kKbDxs+hg5XQ6vhUFLaEeqcWGpzXBwunWtWc/tCyOp5EomkVihWK0mPPY45Lg7/xx7F97Zb8bl5ilrXo2u36geoBkeG0ulbtGgw10erADc0ybvAYoSBpX5nJ3fQu1YQasViUav+1aPbupxRSySSWmE6fBh7UREhb72J54QJAOj8/Wm9cgVaz5rVZL5Y6Pz9sWZmoiiKQzvBnMoopGeEDxSoqe54nRX+5xZQwfVhSUoCmw1DRN2FWs6oJRJJrTDu2AGAa58+Fbbr/PwaJLGkPuj8/VFKSrAXFTlszGKzjaTcYtU/XZAGQnPGNw3gHlRhRm2OiwPUVPm60qyE2t3dvbFNkEgueYzbd2CIikIXEFD9wY2Mzl9N77Zlnh/XXFdOZxaiKBAd4A4FKeAWWDG5xT3wwkIdJYVaIpFcBBSbDeOuXefNppsqWj9/AKwOFOqy0Dx1Rp1aIV4aOM/1YY6NQ+PujraOER/QTIU6JSWFoUOH0r17dzp37syGDRsa2ySJ5JLAdPQo9sJCXPteGkKtCygT6iyHjXk6owghINLPtVSoQyoe4B5YmkZuAcoaBUTWy0feKIuJqe++S8kRx5Y5derQnuCXXqrRsXPnzmXcuHG8/PLL2Gw2jEajQ22RSC5XjNsv7J9uquj8VNeHNctxM+qEbCMhns4467VQmAphvSoe4B4EKPDHHTBIjY6pqstNTWiWUR99+vThrrvuwmKxMGnSpPI61BKJpGoKVq3CqU1rNevvEkDr4wMajUNdH4k5xYT5uqoz5qKM82fUHa5WU8qP/o0SvwtLkhbPKyfW65qNItQ1nfk2FEOHDmX9+vUsWbKEO+64g6eeeorbb7+9UW2SSJo6lpQUinftIuCJxxvblBojtFq0vr7YHOj6SMgxqqVNy8qZnuujdg+Aqz+DtldQ/MXtYA/AKbr2hZjOpln6qOPi4ggKCuLee+/lnnvuYXcDdyqWSC4H8pf9A4Dn+PGNbEnt0Pn5Yc1yjFCXWG2k5psI9yn1T8P5M+oy2o4jLzkQoRO4Dx9Wr+s2S9fH2rVrmTZtGnq9Hnd3d3788cfGNkkiafLkL12Kc6dO9YoHbgzKkl4cQUquCUWBcF9XKFC72ag+6fOxW6zkx+rxCM1FW5wM7m3rfN1mNaMuLFTDav7v//6PgwcPsmfPHjZs2CA7uUgk1WBJSsJ08CCeEy6t2TSosdSOiqNOyFEDD8J8XNQYaqh0Rl24ejV2oxnvNnZY8YraBaaONCuhlkgkdcO4S22J5TZoUL3HKrHa6j1GbdD6nUkjry8J2WrjEHVGnQpCC27+Fzw2b9Ff6EJCcL35BTixHA79WefrSqGWSCTVYtyzB42bW72r4+2Ky6HL6ytYd7zyBrCORufvj2I2Yy/9RV0fEnOM6DSCYE9nVajdg0Bzfl9Ie0kJRVu34jF6NGLAA9CiB/zzYp1n1RdVqB3xjXap0RzvWXL5Ubx7Dy7dutW4WW1l/LEzAbPNzkt/HqCoxOog66qmLI3cEX7qhJxiWni7oNUINYb63IiPUop370YxmXAbOEAV8m43q1Eixrotal40oXZ2diYrK6tZCZeiKGRlZeHs7NzYpkgktaZw4yZib74Fc2wsJceP49KjR73GK7HaWHYwlU4tPEnKLebjlccdZGnV6PxV14Qj/NQJ2UbCfV3UJxdKHy+laPNm0Otx69tX3eCp9m8kv261sS9a1EdYWBiJiYlkZFy8nzxNAWdnZ8IaoAOyRNKQ2E0mUl9/HUtSEomPPgaKgkvP+gn1+uOZ5BVb+HRKdxbsTmLerkRemdjBoeVHy1AUhSMpBXQI8ThT78MBIXqJOcWMah+ott/KOgktLxx2V7hpE67du6Nxc1M3eJQJdTKE1L5m90UTar1eL6MrJJJGxl5aLkHj6lphu2K3Y9yxE9eePRB6Pdk//IAlKQnX3r0x7twJGg0u3erWFMBstbMjNpsfNsXg46pncGt/ErKN/LUvmbT8EoK9HP+Lc/bmWN5YfJi59/Sjb7nro35CXWy2kVlYos6oE7aB1QStzhdqa1YWJYePEPDEE2c2ep4l1HWgWcZRSyTNlcQnnwRFIWLGjArbc375hbS33sapYwdcunUj9495eIwbR/Abr3Nq7Dj0oaFo61AmOM9o4a7ZO9gVlwPAfUNboddqaBekdu8+mprvcKFOyDby3+VqQf+NJzMZMFaNX7ZlZ9dr3MMpeQC0CnCH02tBo4PIgerY+fnYi4rQh4RQuG49AG6DBp452T1QjRCRQi2RSKqj5MQJrOkZ2AoK0HqoYmk3Gsn86msMraOxpqaRe+wPvG+YTOCTT6L18iL866/q1BDAZLFx04wtnM4o4oPruzC0bYAaLQG0D1Y7wRxLLWB4u0DH3SDw+l+HEECrADe2ns5CaDQIZ2fsJlO9xl1xOA29VjC4jT9sWQuhvcFJfQ3T3nmHwnXriV7+Dzlz52Jo2RLnTmf1SNRoVX92Wex1LZFCLZE0ExSbDWtaOthsFG3egue4sVjS0sn5+WdsmZmE/e9TnFq3xl5YiD40tPw81169qhi1cracyuJoagGf3dyDq7u1qLDPy1VPsKczx1IL6nVP53I6o5DVR9N5Zmxbisw2vl1/mqISKxoXF+zF9auSufJwGv1b+eGpFELyHhj2fPm+ou07sOXmkvTkk5gOHiTotVcRmnNiNTxbNP3FRIlE0rhYMzPBpiabFK5fR8mxo2R++RUA7iNGlAuy1svLIdfbeDITJ52GsR0vnGLdLtiDow4W6t92JqDVCG7sE86RlAK+WnuKXXE5tHBxQTEW13nck+mFnM4o4o6BURCzAVDUruOoxaqsKSloPD0p2rwFjYcH3tdcc/4gHiGQUbfyzlKoJZJmgjVF/dmt8fKiYMVK7EVFeIwdi/f11+Hav7/Dr7fxRCZ9onzVus0XoH2wB1tOZWGx2dFr6x8pbLHZmb8riZHtAwn0cMbNoEOrEWyLyeI6V5d6uT5WHlYr5Y3uEAS79qn+5lD1i614714AQt5+i+TnX8Bnyk1noj3OxjMUTq2p0/VlZqJE0kywpKrV3rwnXYO9oACdnx8h77yN+7BhaJycHHqt9AITx9IKGNT6wunVoM6ozTY7sZnnN55VFAWz1Q6o8dezN8disdmx2RWmzNjC238fLt9fxr9H0sgsLOGm3uEAuDnp6BrmxdbT2Wic6+f6WHM0nc6hnrTwdoHibHDxBp0BULM2hbMzHiNG0HrFcgIee+zCg3i2AHMBmPJrfX0p1BJJM6FcqKdMwdCyJSFvv1W+oOhoNp9UQ+GGtKlaqIELuj9+2BTL4A9WY7LY+HtfCq//dYjVR9OJySxi6+lsvtsYw63fbSsX64RsI68sPEgrfzeGtzvTdLdLqBfH0wrQ1MP1YbLY2JuQy8Do0nspzgVn7/L9xbv34NKlC0KvRxcQUPnCaz1C9KRQSyTNBGtKKsLFBUNUFNHLluI+rH41ki/E9phsRny4lveXHcXbVU/HEM9Kj20d6I6TTsOmk+dnDK47nkF6QQlbT2extrQuyKGkPA4lqyFyN/eNYHtsNvsScymx2rhn9k5KrHZm3N4b3VlulDAfFwpMVmxOdY/6OJiUh9lmp3dkaXNaU646o0aNmDEdOVKzrM0yoS6QQi2RSCrBkpqKPijI4ZmAyw+l8u360wDM3RZHer6JAA8n7hgYhUZT+bWcdFqu6xnGgj1JZBWWlG9XFIW9CbnlY284oQr1gaQ8Dqfko9cKHh+lFofaHZfDttPZHEsr4N1ru6idwc8izEdN7DFp9NiL6zaj3hGrxoD3KhPq4lxwUR/n/7McbLaaZW2WlUOtw4xaLiZKJM0ES2oKuhDH9jrMNZp5bt5+8k0WRnUI5N8j6UzoEsK0G2qWxXj34Ch+2R7Pz9vieaxUfGMyi8grtuCk0zB/VxJmmx0fVz0HkvKx2hXaBHoQ7OVMS383dsXlkG00o9MIRrY/Px47zEety1Gk0aOvo496Z2w20QFu+LmX+vGLc8C3FSWnT5P69tu49O6F++DB1Q9ULtS1j6WWM2qJpJlgTUlFH1xJ26g68sXqk+SbLGiE4PFf91JQYuWKzjX/Mmgd6MGIdgH8uCW2vE512Wz6/wZGYbbZEUJ9nFlYwo7YbDq1UN0pPSK82R2fw+aTWfSM8MHN6fx5Z9mMukDoUIpr7/qw2xV2xuXQJ8r3zMZS10fKSy+jcXYm9KOPELoazHn1zuDqD7mxtbZDCrVE0gxQLBasGRnoHTCjVhSF67/aTM+3VjJrcyw39gpnbMcgDiTl4WbQVhnpcSGu7xVGZqGZE2lqvei9Cbm4GbTcO6QVGgHdwrzLFyVNFjsdS4W6V6QPmYVmDiTlMbC13wXH9nHV42rQkmfX1sn1cTKjkLxiC73LhNpuB1Medq0Hxfv34zNlCvqgC8eJX5DIAXBiJdhr1zyhRkIthPAWQswTQhwVQhwRQgyo1VUkEkmjYs3IAEVBF1x/od58KotdcTl0DfNibKcgnh7XltsHRAEwskNQpXHTldG2tO7HyXRVqPfE59I1zJsADyeeHtuOR0e2pmOIF2Xu7k4t1IScnhE+5WNU9uUghCDMx4UcmxbFZEKx2y94XGWUze57RnirG0ryQbFTkmUDux2n9u1qNR6dJ6t1qWM31Oq0mvqo/wf8oyjKZCGEAXCt7gSJRNJ0KAvN0ztAqH/aGoePq56vb+1VLsoB7k48NaYtYzvVYnZZSpSfG1qN4GR6ISaLjSMp+dw7tBUAD49oXX5c60B3jqcV0j5EFfa2QR64O+mwKwrdwrwrHT/Mx5VMqzonVYqLERdKRqmE5Nzi8jEA1e0BlKSp253bt6/xWKrR48DgAQf+KM9srAnVCrUQwgsYCtwBoCiKGTDXzjqJRNKYWEqzEus7o07NM7HicBr3DG5ZYeYshChfDKwtBp2GSD9XTqYXciApD6tdoUe493nHDYz2R6/V4OmsxilrNYJxnYLRCHWMygjzcSHdoj62m0wXzhqshLR8E/7uhjPjF+cCYErKQ7i6oq9trXm9C3S4Eg4vhokfg65miUY1mVG3BDKAH4QQ3YBdwOOKolRIJxJC3AfcBxAREVEr2yUSScNiLZtRh9RvMfGLNSewKwq39HPsZ7x1gDsn0gvYEauWIi0PhTuLlyd2wGav2CHqoxurjy4J83Fhu12Vutr6qVPyTBXLsBaroXol8ek4t2lzfuGlmtBxEuz7BRK2Q8shNTqlJlfRAT2BrxRF6QEUAS+ce5CiKDMURemtKErvgICAc3dLJJJGxJKahsbNrV6ZiH/vT+anrfHcNaglkX41n5XWhNaB7sRlGdl6OptWZ4fCnYVeq6m1/xtUt4WpNN27rHFCTUnNM5WXZgXAlIuigCk2Gafauj3K8Ct159Sikl5NhDoRSFQUZVvp83mowi2RSC4RrPWMoc4uMvP8vP30jPDmhfF1FKgqaB3ojtWusPFExpkMQAcR5uNCiVZ1lyi1zE5MzT93Rp2L1ajBXlCIU7u2dTPIs/aJL9UKtaIoqUCCEKJseXMUcLj21kkkksbCUs8Y6nXH0yky23j9qk4OqXR3LmUZhXaFM6FwDiLMxxWTtmxGXXPXh8liI9doqTijLs7BlKeKfq0XEsswuIGTV8UmAgWpVZ5S01f8UeBnIcR+oDvwbt0slEgkjYElNRVdcM0jMhRFYcOJDKavOYnVZmfDiUx8XPV0CXVMrepziQ44k/rt6Bm1j6sei151pdSmgl5qnjr7DvZyObPRlEtJnircTm3qtnhqsVtY7OOPNe8s18fx5VWeU6PwPEVR9gK962SVRCJpVOxmM7bMzBrPqK02O7d/v53Np9QKeOG+rmw4kcngNgFV1u6oD25OOkK9XTBZbLT0d6z/WwgBLqrY1sb1kZpfKtSeFV0fxgxXnNq2rbO/f/7x+bzjYsGtKI6RZRvTq3ZSyMxEieQyx5qeDlDjrMR/j6az+VQWT49pS0t/N95dcoSMgpIqS5Y6gqu6teCmPuEOLxoFoC0V6tq4Ps7MqJ0xJyRgTkzEnp+FMU3gNqB2jRaOZR/jm33fYLFb+PnIzwActOSeOSDtUJXny6JMEslljrWaGGq7XWH2lljmbI3jwxu68dPWOEK8nHlweDTebgZeXXgQqLq2tCNoiEXKMrSupUJdi/C8shm1f3EusTfegMbdneAhCoqNWnfE+eXoL8w/MZ/NyZuJzY9Fh2C/MKup5EIDaQerPF8KtURymWM5K4a6xGpDUcBZr8VksbFwTxK/7khgb0IuBp2G+37cSWahmafGtEWn1TC5ZxgfrziGr5uBkLN9tZcYOlc1s1Ax1W5G7aUX5L78AraCAmx5eaSv1IIA1z59qj0/qzgLIQS+zr7sz9yPQWNgd/puAl0CGWzwZ4X9IPbCNDSI8vjsSu2vsdUSieSSxJJSKtRBQdw1ZxeZhSXMf3AgT/y6l38OpdLS3433r+tC1zBvrvtqEzqNYEoftZ2Vi0HL17f2qlCM/1LE4Kr6mWvj+shLTuO1rd9TnHCIFh+8T+aXX1ESF4dLhCdad/cqz7Urdu5ZcQ9uejdmjJnBqdxT3NPlHuyKnS7+XciLXc+feYeJTd1NK+Fc5VgghVoiueyxpqodsoWrK7vjcsg3Wbnpm63sTcjl2XHteGh4dLlf+Ic7+pKWbyLwrAW0fq0uXJnuUsLV2YBZZ6ix60OxWJj8/et4FuUS/MbreF1zDfZiE6lvvIFb++p9/aviVnEy9yQAqxNWY1fsdAvoxtCwoQCcLC6EmPkcTNtNK1F5F5wyLu2vSYlEUi2W1DT0wcGk5ZeQb7IS4uXM3oRc+kb58sCw6AqLdwOi/ZjUI7QRrW0YXA1aSnSGGofnGXfvwTc/kw3XPYTPlCkAeF1zFb7tCvEe1qXKcxVFYcb+Gfg6q/HgX+79EoAu/mfOaxncE1e7nQM5xyDtEDHeVb/mckYtkVzmlHV2OZqqdr9+//quHEjM5fpeYWgbKNyuqeFm0FGiNdS4eUD26jVYhBZL3zMVnTWKiaAe+RBadSGmX4/9yrGcY7w16C1mHphJbH4s4R7h+DifiQ/XeoTQqcTCnsJ4lDwbs3yrTvKRM2qJ5DKnrLPL8TS123fXUC8eGdnmkl4crC1uTjqKtTV3fWT+u5ZDfi0Z2i3yzMayBb/SxrYXYsb+Gby77V0GhQ5iYquJjIgYAVScTQOg0TLapuOYJZefTYn8peRXaY8UaonkMqbkdAy2nBwMEeEcSy0k0MMJHzdDY5t10XFz0mLU6qt1fSiKgiUlBefEWE636kb3s8utlpY4LWtsey77Mvbx+Z7PmdByAp+P/By9Rs+oiFEAdAs4v8rfjfpA2peY+cDXC6Wa2HEp1BLJZUzm118hnJ3xmjSJ42kFtAuue/W8SxlXg44SjR6bsXLXx/G0Anq+tZI/vvwdgBZXjKqYfGMqnVE7e593rqIofLrrU3ydfXl9wOvoNWo9kK7+Xfl85Odc1+a6887ReYbyalY2AhjfckKV9ksftURymVISE0P+30vwveMOhI8vx9MKuLV/ZPUnXoa4OWkx6QxYqyhzuvpoOiIrkxbrfyPJzZ8x4/tVPKB8Ru1dYXNmcSar4laxM20nL/Z9EVf9mQZYQgiGhw+/8AV73EbXoE781mUSUd4teZ/3K7VNCrVEchmRt2gRGi8vXHv1IuXVVxEGA35330V8tpESq71Zz6iztAZsxsp9wXuPJfPf7TPxs5vY9vBbjPY5p+NguY/6jOvjcNZhblt6G2a7mWivaG5oe0PNjWp3BbS7gg41OFQKtURymZC7YCEpL74IgMbLC7vRSIv33kPn58exg2rSS7ug5inUbgYtSTpDpUWZ7HYFsWUjoTnJhH05nS4jR55/UGm/xLNdH0tOL0FB4ftx39PFvwv60rrXjkYKtURyGVC8fz+pr72Ga//+uA8fRt7CRQQ99yxuAwcCcCpD7fAdHVh1Rt3liqtTWXjehaM+TmYUEpV6EpuTM+5Dh154kOJc0LtCabcYRVFYHb+afiH96BNcfUp5fZBCLZE0cWz5+Wg9K89esxcXk/zsc+gCAgj95GN0Pj743XFHhWNiM4sI8HDC3al5fuTdDFq1eUBprQ+72UzJiRM4t2+P0GrZEZtNp+xY9J27InSVvEbFuRXcHidyT5BYmMhdXe5qcPtl1IdE0oTJ/2c5x/sPoHj//kqPyfjfZ5jj4gh59x10PhcOHYvNKqKlg/scXkq4GnSYdAaE2Uz6hx9yvG8/Yq+fTNp772O3K+w+kkjL/BR8+ldRdt+UW8HtsTp+NQLBiPARDW5/8/x6lUguAWz5+aS+/TbY7eQvWYpL167nHVNyOobs2bPxnnITblWU3ozJNDKiXfNtOu3mpC3vm5j13Uw8xo1DGAzk/PQTL57QkKd1QqMouPasoh1scW55xIeiKKyMW0nXgK74uzRs+VeQM2qJpMmS/vHH2LKzMbSOpmDVKhRFOe+YvAULQKPB/6GHKh2nwGQhs7CEKAd3TrmUcHPSlfdNdO7aldAPpxHy9lucDmzF47t+5ZmCPaDR4NKte+WDFOeUuz7WJa7jeM7xC8ZHNwRSqCWSJoglLZ3cP+bhM2UKfnfcgSUpiZIjR1BstnLBVmw28hYtwn3wYPSBgZWOFZelxg47usXVpYSbQUesZwjFASGE/vcDhF7P1oQCXul5KwQF43NgJ07t26F1r+I1KnV92BU7X+z5ggiPCK6Kvuqi2C9dHxJJEyTvz/lgs+F7+21oPD1B8zqpb79DydGjaLy88BgxAqc2rbGmp+P18stVjhWbVQRAVDP2UTvrNRwOaMXaG7+iZ1QUAD9vi0fx9aPNz3PIfP45PEZdICTvbEpdHyviVnAs5xjvDXmvPAOxoZFCLZE0MRSbjZzf/8Bt4AAMpaLi2rs3xu3bcRsyBI2zE7nz5qGYzWi9vHAfMbzK8WIzS4Xa37XK4y5nhBC4GXQUmW0AbD2dxfJDqdw5KAq34EDcZs+qegCrGSxFmAzufLLzE9r6tGV81PiGN7wUKdQSSROjcP16rCkpBL3wQvm2kLffwpKYiOuAAQghsGZkkP3TzzhFt0JjqLrIUkymkSBPJ1wNzfvj7mrQYjRbWXEolUfm7iHK3417h7aq2cmlyS4/GE+RXJTM94O/R6vRNpyx59C83zmJpAmSt3ARWj8/PEaeCfsyRERgiIgof64LCCDwyScqHcNstfPu0iOk5BUTl2Vs1m6PMtycdBSV2Ph01Qmi/F35/f4BeLvWrJLgy1vfYmlUONbMbYyLGtfgCS7nIoVaImlEFEXBnpeH1tsbAFthEYVr1+I9eTJCX3v/p92usD02mw+XH2NnXA46jcBqV8p7IDZnXA1aCkwWTmUUclv/yBqLdGZxJkuS19PHZKJn22u5uV/VawINgRRqiaQRyf/7b1JeeZXolSvQBwZSuGY1SkkJnhNq7/+02xWmzNjK9thsPJx0fH5zDwI8nHjyt70MbN3wsb5NHTeDjuNphZRY7bSuRSr90tNLsSl2XszKoVXbKRcsc9rQSKGWSBqRgtWqMBfv2oV+/Hjyly5DFxyMS48etR5rxeE0tsdm88zYttw9uBUuBtWHuuXFUY42+5LE1UlLUq6aQl4bof7r1F90dm1BK0t8ld1dGhIZRy2RNBKK3Y5x23YAjHv2YMvLo3DjRjzHj0doavfRVBSF6WtOEuXnyoPDW5eLtOQMbmfVOampUB/LPsaxnGNc7d5a3dAIs2mQQi2RNBolJ05iy84GjYbi3XsoWLUKLJZauz0UReGPXYkcSMrjweHRzaZhbW1xK/3y8nc31Ng/vS1lGwCj9aWuI2evBrGtOqTrQyJpJIxbtwDgeeVE8pcsJXf+n+jDw3Hu3LnGY5zOKOSFPw+wPSabDiGeXNuj6g7ZzZmy8MTogJq7PWLzY/Fy8sLfbAInT9A2jmTKGbVE0kgUbd2GPjICr4kTwWajePduPCdMqNinrwoW7U1i4mcbOZ5WwNuTOvPXI4Mw6ORHujLcnNQZdW1qcsfkxRDlGYU4p3LexUa+qxJJI6CYzRh37MCtX39cuncv3+45oeomp2UcTyvg2T/20znUk+VPDOXW/pHotfLjXBVlM+rWtZxRR3lGqQkvjbSQCNL1IZE0CgXr1mEvLMRj1Ei0Xl44tWmDothxatsGAKvNzoM/76bYbOPKriHc1Ce8fKZttdl59o99uDvr+OrWXvi7OzXmrVwylPmoq1pItCt2NEL9wiswF5BZnElLr5ZQtAVcfS+KnRdCfgVLJI1A3p8L0AUE4DZoEAAtPvyQsE8/LRfjHzbFsvJwGjGZRbzw5wFWHk4rP3f+7kT2Jebx5jWdpEjXggg/V5z1GjqEXLhbzrHsY/Sf25+96XsBiM2LBSDKKwryEsCr8ZKGaizUQgitEGKPEOLvhjRIIrncsWZkULh+PV6Trilv++Tcri1OrdUQsPgsIx+tPMboDkGsfXY4ns46lh86S6h3JdE60J2JXUIaxf5LlRHtAtn5yhgCPC785bYmYQ3F1mK+2PMFoLo9AFq6hkJhGnhHXixTz6M2M+rHgSMNZYhEcjljOn4ca2YmALnz/wSbDa9rL1x0/su1JxEI3prUCb1Ww6gOQfx7NA2rzU5CtpHtsdlc2yO0xouOEhUhRJU9I3ek7kAjNGxL3cbO1J3E5MWgFVrC7aUNG7wjKj23oamRUAshwoCJwHcNa45EculTsHo1hevWlT8v3reP2Osnc/raa8maOZOML77AbcgQnFq1PO9ck8XGkgMpjO8cTIiXCwBjOwaRa7SwIzaHRXuTALime4uLczPNhBJbCXvT93JD2xvwd/Hn092fciL3BGEeYejzk9WDGlGoa7qY+CnwHODRcKZIJJc+Rdu2k/joY6AohH3+GU7R0SQ+9ji6ALVfYfq0D9VWUJ98DKjCHJdlpF2w+tFaeyydApOVa3qElo85tG0ABp2Gbzec5nhaAX1b+hLm03xrSzcE+9L3YbabGRI6hG4B3Xhp40sADA8bDrnx6kHejeejrlaohRBXAumKouwSQgyv4rj7gPsAIiIa75tHImksLGlpJD35JIbISDTu7qpg2+0IFxei5v6Mzt+f3Pl/4nPzFDRubvy0NY7P/j1BekEJn9zUjWt7hLFwTzL+7k4MivYrH9fNScewtgGsPJyGm0HLW5NqnhAjqRnbUrehERp6BvXEw+BBcmEyX+z9Qo34yI0HjQ48Gm9NoCYz6kHA1UKICYAz4CmE+ElRlFvPPkhRlBnADIDevXuf34VTIrnMyfj4Y+xFRUTO+RGtry/p/52GU3QrPK4YjyFMnSH7P3A/AB+tOMbnq0/St6UvEb6uPD/vAPsS8lh9NJ2p/SPQnRMT/cH1XXlkhJGOLTxlvHQDsCN1B538OuFhUH/Z3Nf1PsI9wukZ1BOWvQxeYXARGwWcS7VCrSjKi8CLAKUz6mfOFWmJpLljOnaMvL8W43f3XThFRwPQ4r13ATiUnMcffx3C393AiPaB/LQ1nl+2xzOlTzjvXtuFfJOF677azE9b4+gd5cM9Q87vOuLrZsDXrWb1KSS1I92Yzt70vdzf7f7ybUIIJrQqTT7KjW9U/zTIhBeJxCFkfPIpGg8P/O69t8L27zac5u0lRzDoNJitdj5ccRydRnDnoChemdgRjUbg7WpgyaNDsCtKhQpvkovDsphlKChMaFlJVmheArRu3FKxtfpfoSjKWmBtg1gikVyiWNLTKVy7Fv9HHkHr5cXHK4+jFYLbBkTy6aoTDG0bwOdTepBRWMKmk5mM7hhEqLdLhTFkWVLHkm3K5su9X5Jfkk+3wG5M7TC10mOXnF5CJ79Oqj/6XKwlUJDSqDHUIGfUEkm9Kd61CwD3YUNZtDeJz/49AcDSAykYzVZendgBL1c9Xq76WhWsl9QNRVF4eePLbE3Ziofeg3/j/2Vy28k4ac9PdDmde5oj2Ud4rs9zFx4sL1H928iuD7kqIZHUE+OOnQhXVxL9wnhlwUF6RfowoUswx9IKuK5nGG2CZFTrxWTu0blsTNrIc32e4z8D/4PZbmZ/xv7zjrMrdqbvnY5WaBnfspIa4Llx6t9GTB8HOaOWSOqNcdcujG06cvs32zHoNHx6U3cCPJzoGRHHtWfFQ0sangxjBv/b/T+GhA5hSrspFFgK0AgN21O3n9c5fPre6ayIW8ETPZ/A36WSnpJ5aoIRXo37PsoZtURSD2y5uZQcP848iz/RAW4sfnQw4b6uOOu13DOkFX6yaNJF5ct9X2KxW3ix74sIIfA0eNLetz07UndUOO5o9lFm7J/BdW2u467Od1U+oClP/evi04BWV48UaomkHhh37wFF4YBvSz66sdt5i4SSi8fpvNMsOLGAG9veSLjnGVdFn6A+7M/Yj8lqKt/2x7E/cNI68VSvp6qumWLKAwQYGtd9JYVaIqkDKXsOsmPgMJJefQ2rRofo0InWgdIXfbHIK8ljU9ImFEXNrUs3pvP46sdx1blWiIcG6BvSF4vdwt6MvQAYLUaWxCxhXNQ4vJyq6YFYkq+24Kpls2FHI4VaIqkFZcKw97+foc3L5YhLIPNaD2VC76jGNewyxmKzsPT0Uqx2KwDLY5dz9cKreWDVA3y972tO5pzkruV3kW5M58vRX+LrXLHAf8/AnggEe9L2AGrcdJGliBva3lD9xU35jdbQ9mzkYqJEUkN+mf47ectX0OvZRwjdu5nNnYfxfrSaJLGpm6xmVxMURWFH6g4SChKI9Iykd3Dvas+ZdWgWn+35DCEEvYJ68dz65+jg24HeQb35ct+XfHvgWzwMHnwz5hu6B3Y/73x3gzthHmGcyFXDJpfELKGVVyu6BXSr3mBTHjhfuNHAxUQKtURSA07sPUabr97DzWrC+OBOtIqdMa88Brk6UvNM0jddQ34+8jMf7PgAAA+9B+umrEOv0Vd6fGZxJt8dUKsrr4hdQbYpG7ti590h7xLhEYFhk4FiazGv9H+l8sgNINo7mtO5p1EUhSNZR5jYamLN6nmXuT4aGSnUEkk1FBcVE/vk0/gKKLjvcZy/+4K4Dr2Z0K0d0Y1t3CVEQkECn+35jEGhgxgbOZbXN7/O7rTd9Avpd8HjLXYLn+z6BLPNzKDQQWxI2kBSYRJtfNrQykuth/LekPdqdO1or2g2Jm4kLj+OQksh7Xzb1cxoUy54htXs2AZECrVEcg4n0wuYsyWOraez8XXWMG7+5/RLOcWph17iysduI2fSeNr5N16j00sRu2LnP5v/g0ZoeGPAG3gaPHln6zusTVh7QaE+lHmI5zc8T1x+HHd2upMhYUPYlLSJI9lHeKT7I7W+frR3NFbFyoq4FQC092lfsxNN+RDY+DNquZgoafYoikJhibpQtfFEJuP/t4FfdiQQ7iK4buk39EvcT/F9j3HlY7cB4NMqEidPGeFRxncHvuPhfx/GYrdUeszsQ7PZlrqNZ3o/Q7BbMK56V/qF9GNd4jrMNjMxeTHlxxZZinh63dOU2EqYPmo6T/Z6kp6BPfFzVmt0j40aW2sbo73V3z5LTy9FIzS09mldsxOl60MiaXy+3xjDnK1xxGQWMTDaj30JuUQHuDN7QhhFzzxJyckTBD73HH533dnYpjZJ8krymLF/BsXWYmYemMkD3R6osD+lMIUNSRv4bPdnjI4YzfVtri/fNzx8OG9tfYvJiycTlx/H4kmLifCMYNqOaaQUpTD7itnli4NaoeXm9jdzMPPghYsnVUNLr5YIBKfyTtHSqyUuuhqsKSiKjPqQSBqb9cczePPvw/SO9GFce38W7E/F29XAzNHB5N9/D/aiIsJnfIP7kCGNbWqTZf6J+RRbi+kR2INv9n3DkNAhdPLvBMCa+DU8ufZJbIqNaK9o3hj4RoUFvKFhQwFIK0rDrtjZmrIVq93K/BPzuaPTHedFcJwbH10bXHQuhLqHkliYWHO3h7kIFJuM+pBIGguj2crr83bzSPwaJu07jDUxkXuefRZ9ly6kP3gPitVK5OxZOHfo0NimNlksNgtzj8ylb3BfPhr2EZMXT+belfcybeg0jFYjL254kQ6+HXhr0FtEe0efF2UR7BbMN2O+IcozituX3c721O3km/MBuLWD43uTtPZuTWJhIm1929bshBLVFun6kEgaie/mruX5Be8RWZCGYeBA9MFBZLz7Luh06IOCiPjhe5xa19CP2QxZcGIBX+37ijRjGq8NeA1vZ29mj5/N/Svv54FVqvsj0jOS6aOnn5eAcjYDWwwEoG9wXzYlbyK5MJlOfp0IcgtyuM3R3tGsTVxLe9+aLiSW1vmQrg+J5OJzMCmPwO8+JchaRPi33+I+ZDCK1Ur6tGmYExIJefstdL4yqqMylsUs47XNr9E1oCuv9H+l3IUR6h7KnPFzWJOwhmDXYLoHdsdVX7Nu6X2C+7D49GKyTdl1iuqoCX2D+zL/xHw6+9WwObCpdEYtXR8SScOTmZrJjt+XMvbhW1GAGV/M58HMU3g+/SzuQwYDIHQ6gl58sXENvQQ4mHmQ1za9Rs/Annw39jv02orJKj7OPlzX5rpaj3t2iN7w8OH1NfOCDAwdyIYpG2p+QtmMurp6IBcBKdSSy55tr/+XVusWs8BYQkyPIQzYsACrty8ht93S2KZdUmxL2cbjax7Hz8WPj4d/fJ5I14cW7i0IdVdrPrf1qaEPuaEp81FL14dE0rDYCgsJ3rIKgNBfZnB8ww66Z54k6MUX0Dg7N7J1lw4rYlfwwoYXiPSM5KvRX+Hn4ufwa7w24DU0QlOz1O6LQbmPWro+JJIG5cTcebiaizk+5UFa//Y1157agMcNN+IztfJmpxKVzOJMlp5eSnxBPL8f+51uAd34YtQX1ZcGrSNlC4tNhnLXhxRqiaTBMCcmUfjTTyR7h9L/yXsRHUPAaiPg5puazqytCVJiK+HfuH95f/v75JTkYNAYGBs1lrcGvVWzRJHLhZJ80OhB3/j3LIVactlhSU4m8YuvKF60AAOCrRMe4movF7ixBvWHmyl2xc7O1J3MOzGPtQlrKbYW08G3A9+N+4423m2a5xebKV91ezSBe5dCLblsyF+xgrz5f1KwaRNWOyyL7MfvbUfx7A2DG9u0Jkm6MZ2lp5eSWZzJ6oTVJBQk4GHwYGKriYwIH8GAFgOqLEF62ZG0C367DW5fBP5tVNdHE3B7gBRqyWWA3Wwm7Z13yf3tN4w+AayMHsLGziN49e6RPBDsiZdrMxKbGpJvzueeFfcQkxeDXqOna0BXHuz2IGMix+Csa4aLrIoCy1+G/CQ48hcMeVp1fTSBiA+QQi25DEh97XXyFi5kYYdRfN9uHKM6tWDmNZ0I8myGglMDiixFPLvuWRIKEvh+3Pf0DurdPF0bZ3NsGcRvAY0OTq1RhbqJdHcBKdSSS5z8FSvIW7iQpT0m8EfHK1j54ECi/N3qPqDVDCtegY7XQNQgxxnaBFAUhX/j/+W97e+RbkznzYFv0ie4T2Ob1fjYrLDqDfBrDW3GwfYZUFKo+qj9m0YZAVmPWnLJYsvLI/nV10kMiOSbyOFMv6Vn/UQaYMOHsP0b+PUWyI6p/vhLhGxTNo+ufpQn1z6Jt5M3c8bP4do21za2WU2DvT9D5jEY9Tq0HQt2C8RtKq1F3TRcH1KoJZcsyb/+gZKXy2fdJzP99r4MiK5nEkbKPtjwEbQeAyjw+21gq7wYflOnxFaC1W5FURRe2/QaW5K38EzvZ/jtyt8u2AS2WWI2wtr3IKwvdLgKwvuDzgUO/wXFudL1IZHUB8VuJ2PuL5z2jeLNZ6+nR4RP/QZM2AG/3gyufnDdDIhZD3/8HxxeBF0mO8boBkZRFL7a9xXrEteRkJ9AgaUAfxd/RoaPZF3iOp7v8zy3dnR8+dBLmj1zoCAFJn+vhuHpnSFqMOz9Sd3v7vgqfnVBCrXkkqRoyxZc0pLZO+puJtdHpI+vgF0/wMl/wTMEbvkdXH2hw9XgGw3bvrlkhPrr/V/z1b6v6BnYk4mtJuLv4s/qhNX8fvx3egb25JYOsrbJeSTuUJvXRp6VFTn+A4iZCN7hENU0mkZIoZZcUlhzcihYsZK0738gz+BG5HVX1n2wkkL47VZ1Ft3jVhjxMriVuk80Guh7H/zzvBpfG9oLkvfAiZUw+EkoylB9mwMeqTJzLbM4k11puxjQYgAuWhf2Z+5nwYkFGK1GrmtzHQcyD3Ak6wjvDn4Xd4N7+Xl2xU66MR0/Z79Kix8pisL6xPXMPTqXQnMh+zP3c3X01bw96O3yKI57utzD6oTV9AzsiUZIT+d5pB+FwHPqU/tFq/+aEFKoJZcE1uxsUt/4DwWrV4PVSmFgCz7vPpmPekfVfdBTq8FWAtd9Ay2Hnr+/+y2w+i1Y/yFM/gHm3QXZpyFhO2Qeh9w48AxVj0MVznRjOgoKAS4BaISG59c/z/bU7eg0OlDAqlhx07th0BhYGbey/FLfH/yex3o+BsA/Mf/w+ubXMVqNhHuE82r/VxnQYkD5sWV9CjcmbeR03mlC3UMJ9wjnxrY38nzf5yuE2mk1WsZEjqn7a3Q5Y7ep72OrYY1tSbVIoZY0eYr37iXxqacxZ2SSPOIqNkT05MdMZ0a0DyTYqx6x0seWgbM3RFRSDMjZE4Y+C6tehx+uUEW6x62w5ydw8VFn4seWQfdbyCrO4o3Nb7A2cS0AbXzacH2b69meup27Wl6DcPNHIzS09W3L0NChaISGDUkbaOXVim8PfMuPh3/kxnY34ufsx7Sd0wj1CGVS9CR+P/479628jy9HfcmQMPVn+LQd0/j79N/0De7L7R1v5+rWVzevDEJHkR2jflEHNv12a0JRlKoPECIc+BEIArXuuqIo/6vqnN69eys7d+50mJGS5oklPZ2Ul1+haMMGcl29eL3P/3HcJwI3g5a7h7Ti3iEt8XCuo0DZrPBhG2gzRl08rAxFUWfSh/6E9lfClJ8hdiOxGvhy0xtclXwC+/Xf8dq2tyk0F3J3l7vxMHgwfe90iixFtNN58tvJI2hvnQ/RI9RZnEZb4RLJhclcteAq+ob0ZUjoEN7b/h5fj/6aQaGDMFlN3LD4Bqx2KwsnLSSxIJHr/rqO2zrcxjN9nqnbvUtUjixWXV/3rIawXo1tDUKIXYqi9L7QvprMqK3A04qi7BZCeAC7hBArFUU57FArJZKzsObkkHD33RTHJzKn0wT29x7Nc9f0ZEC0H856LVpNHTPpbFY1TtqUB8XZ0G581ccLAddMxxbUiZ/cndm5+lFuancTb299myRzGsv8PWHdU7T1act3Y7+jjU8bQG379N8d/+WJU3vRKjb44w4I7almvfW+C8b8B5w8ALVo/nN9nuOdbe+wMWkjHXw7lJf8dNY583L/l7l3xb28tuk10oxpuOhcuLvL3XW7f8kZ0o+qfwPaNa4dNaBaoVYUJQVIKX1cIIQ4AoQCUqglDqf40CFyfpxD0fbtlGRk8lK/u3Ht15ffbu9d99nz2ez7BZa/pD7WuUDr0dWecrQwnveKDrA7ZjcuOhfWJqzFWevM7NHfcvSPW8hv0YW7Jv6CQWsoP6edbztm9v8PbOmkLkoenA+pB6DTtbDze3WB8t416qIlcFP7m/B38eftbW/zaI9HK/iZ+4f05/o21zP/xHwAHu/5OD7O9QxHlEDGEfCKACf36o9tZKp1fVQ4WIgoYD3QWVGU/HP23QfcBxAREdErLi7OgWZKmgNF27cTf/+DlAgNsUGtmNViAJFjhvHB9V1x1murH6A6bBb4vKfqW570FQgNBLRDURRSi1IJcQ+hxFbCtB3TOJR5iNySXAxaA6fzTuNp8OS5Ps8xKmIUvx77la7+Xekb0hf+uFMt4jPgYRj+YsUIkN1z4K9H4MEt4NkCdM5qnO6en2HRQ3Djj2qqeg1QFIUCSwGKouBp8JS1ORzBlwPBKxSm/tHYlgD1d32UDeIOzAeeOFekARRFmQHMANVHXUdbJc2IxBPxbH//f/gc2o1fQSZ6m5UEjyBeG/oAXmEtuKF3GHcPbll/UUreCytfVRcAc+NhwkflC0h5JXn8Z8t/WBm3kqkdppJtymZZzDL6h/QnwjOCElsJYyLHcHun2/E0qFlq93S558zYEz8Cgyts+p869uQfYOdMdV/sRnAPVq919j10mwIbP4Z109R47RrcnxCi/PoSB2CzQtYJaD2qsS2pETUSaiGEHlWkf1YU5c+GNeniY9y9m8L16xEaLd433Yg+qGlkI13OpB48Tuwdd9LamEdCZEf2hnUmxa4ndfA4/rp3GIEeDqx8t/tHVTQVRU0VbqOGq+1J38Pz658nw5jB4NDB/HzkZ0B1LVQQ46pw9YVrpoNPSzWUz2yEE8vP7O928/lCrNHCkGdg4QNw/J/q/eR1xW5Xr13bLzqbRa110moEDHioYWy7GCgKLHtejX+/5Tf1vSpjx3dgM0OL7o1mXm2oVqiFOp2ZCRxRFOXjhjfp4pI7fz4pr78BNhsoCkXbthE550eERiYHNAR5f/1F8uyfMB89ik7nhPWzb5k4egDjbHb2JuTSNcwbg86Br72iwMlV0PYKuPpz1f0gBL8e/ZX3t79PsFsws8fPpmtAVxadXERmcSZ3db6r9tcZ/KQal31iOXS5QV2gWv2OWj/iQnS5QU2mObbM8UJtyofVb8PhheAdCfesrPzYnDjVf96iO0SPVLedXAUnVqj/itLVYkWXoqtl1w/qwjHAnEnwf4vV+tLHl8PyF9Uono6TGtPCGlOTGfUg4DbggBBib+m2lxRFWdpgVl0EFEUh8/PPyfzyK9wGDSL0f59SsHw5KS+/Qt6ff+I92fFpw4Xr1mHLzcXrmpr5JS83suf8RNo77xDjFcLRNoPo99T99BzaHQCdVkPvKN+qB6jTRU+riSkDHwU3fwBmH5rNhzs/ZFjYMN4b8h4eBjX64prW9XhfNFq4YRYcWwrdbwWtDvo9UB7ZcR5anZrtmLSr7tdc8KDqZ580/cw2ux3+vFfNoPRrDYnboSAVPIIrnluYrpZz3f87atQtqo992PPqgqurP3S4EjZ+Av7tIKI//PWo+guhRw0bA5cUwoqXVXFsObRGC7cOI/2oOptuPRr63Au/TIFtM2DYs7DuA/W1uW7GeaGSTZWaRH1sBC7Br9PzUex2hEaDYjaT8uqr5C36C6/rryPkjTcQej1e111H7oIFpP13Glp/fzyGD3fIde0mE9lz5pDxUekPEo0Gr6sqmWldAhSVWNmXmEtqnglnvZaB0X54uxqqPCdr5vekT5vGlhZd2HDzE3wytXe159SL3AQ1/C5+G/+6uvBdyjLuiW/F7rTd/Hj4R8ZGjuX9oe87NlHEPRB63XHmeWUiXUZob7WsakmhWg/Z4Apj3qzZtYqyYP9voNjUhcygjur29dNUd8r4/0J4X5gxXHX7nF2vJHkP/DgJLEYY9Dj0vF09b+17YMyCY/9A7zth3LuQcRyWPacuwObEQuwGdfy2V6guJBcfWPOuKvxXfQpnp7v/+x/YNQu0BtWHP/Zt9QvzYrD+v+p1r52hlgXwDoeMo+ovrMyT6uthqGdJ3IvIZZ+ZqJjNpL7zLgWr/8WWlY3HqFHY8vMxbttGwOOP4ffAA+WLVUIIWrz7LokPP0LiAw+ij4gARUGxWdG6uePaty/mmNMU792H58QJ+EydilPr1gjdmZfRmpODOSYW586dsOfnk/Tssxi37wCbDc8J47FmZJLy0svoQ8Nw7dmjsV6WWpOaZ+LXHfFsO5pK8rHTROYkMSJxN0a7lXujh2Ds0I3e7UJ4YXwHXAxnZinWnByyv/+erG+/43D7fnzW5UZW3NSzYUXalE/67AmkGtPRuvrzYmAAlvwYnljzBAA3t7+Z5/o8p6Z1NyZhfUCxqy6TXT+owjHyNXW2XR1HF6sirdGrs97rv1XdGOv/q7pV+t6nju3sBTHroPP1kHZIdWXMv0fNurxnldobENQoGJ2TWjQf1AVPjRau/Qq+GqTOyu9cBqfXwtYv1UgXjR58IiHrpHqOYiuNphEQu0kdq98D6pfPgvvVGfypNRDUSXUNRQ4E31aOf10zT8DBP2HwE2dqt/hGQ/YpKM6BkryGuW4DclkLtb24mKSnnqZwzRo8J05E6+ND3qJF2IuLCXn/PbwnTTrvHENEBFHz55E9azYlR4+CTovQ6rCmp5M7bx66gADchw8nb+Eicv+Yh3BywqldO3SBAZQcOYolKQkAXYsQBAJrTg5+d9+NS/duuA8fji0vj9gpU0h85BGifv8dnb8fQqerIPZNjcPJ+bw17Xeu3LWYF7Ni0JfVaA4IxK5Ary0zYQsU6F3457uODH/uQXyGDCLnl19Ie/c9FIuFI92G8mzkRF67ogOBDdgiK9eUy+8LpvKtpx2TdyAAgRonfrp2Mf/G/4u7wZ1JrSc12PVrRWhpNty/b4LdqibhJO9WZ8LVcWiBKjbtJqjCOfhJ9a/QqsIohPo4crBasnXNu6qIA7gFwG0LKxYeEkKNhjHlq7PjkO7qdp8o+L+/VBdLix4QOUB1j6QfOrNIe913qotp7btgLoR2E+GfF1T/+KjX1C+A674DrzC1SmHsBnUhzyMEnjzkGPeDopzxo2/4SF2L6P/wmf1+0bD/D8g6deb5JUSt4qhrSmOnkNuNRnJ++52smTOxZWYS/Ppr+Nx8MwC2wkJsOTkYwsMB2JeQy/ebYojPNtI20IOHRkQT6Xfhn0SK1QpaLUIILOnpGLdswXTkKKYjR7Cmp+PUrh0uXTqjCwwi+6c5WJKSCf9yOi5du1YYp+R0DLFTppTb6j5sGOHTv2jAV6RuFGzcxJrDKXx4wsa0lR/i6e6C78TxOHfqhCEiAucuXcBmo2D1Gszx8RzfdRjbts34mIsIfvwx0v/3P/YGtGFGxytJ9gvl4eGteWRk67pnFVZBXkke/1n5MKsz92ETMMYlnDG9HubAkXlc3e0e2oc30bZa/+sOOTFqqc38JBj+AnS9SfV1+7eFlsNAd86vj6JMNf198JPqzPmrgWq4mbkQ+t6rluksY+tXqmgKjRoK2O1mdeHwXJ/12ZwtejVFUWDTp7D2fbCaILir6rO/kCDabWq9lMWPwR1L69/yrCAVZo5RZ+++0fDLTTDwMRj71pljtkxXE53GvacuJD68vcllJFYVR31ZCbW9qIicX34h6/sfsGVn4zqgPwGPPIJrr/Pz+ItKrLz192F+3ZGAh5OODi08OZiUR4SvK4seGYSTrv7f8orNhtBeeJyibdvJ+vZbsNso2ryFVkuX4NSq4X6OfbH6BF4uem4bEFWj4/M2bSbh3vvQ2m0Ynd1wFXZazZ9XrY1f/72H8DefplV+Csnu/sy6/Q1G945mRLtAIvxcHXAnFck357MjdQcfb/uA5KJkbisRjG87mQ5DX66ZC6GxmX8vHPgdhj4Hp/5VBc9mgbQD6v6uU9TqfmdTlkjzwEYI7qK6PH6bqhYZenRXRRFOO6QKuXsQPLxN9Sk3JNkxcHoNdLtFTe6pjJJCmNZaLXI18cMLH2Oz1uw93PaN6kcH0LuqXw53r6p4/WP/qAIePVJ1v7ycWrV9jYBDEl6aMsV795L312Lyly7FlpuL2+DB+D/0UAUf8MGkPBbsSaJdkLrA89nqEyTlFvPg8GgeHtEadycdq4+mcdesnXy4/BgvTehQ70SLykQawK1fX9z69cWalcXJESPJnjOHkNdfr9f1KuOPnQl8uOI4Wo2gR4QPnUOr7gNXfOwYsQ8/SrKbP5qxE2i5bjHBr7xcoy+Su8d349b9TzJw7e+s6DSKWXcOJsDDyVG3UoHNJxbz+KaXMAnwQ8v3WUX0uH+buqh3qRA5AA78cWaxr8w9cd23alfsXbNh5MvgHXHmnNgN4BYIQZ3V5z6RamGh4hzwOCcHIKAD9LhNHb+hRRrAt6X6rzqc3NX+hIcXqb8AznV/LH0W9v2m7us2peoZ/uG/1MiUwPZweh3cMPt8ES6b2cduUl0wTUykq+OSEWrFbqdwzRryl/2DcHbC84rxWNPTyVu4EOP27QhnZ9yHD8fvzjtw6dat/Lzk3GL+s/gQyw+loRFgL/0B0TnUk49u6Ea/Vmf67I1sH8Qt/SL4dkMMu+JymNAlBDcnHeuPZ1BgsvL46Db0cXAImc7PD88rryRv4SICn3gCrZdjm2keS81n3oz5vGSMZ49LMC8t8GTBQ4MqdT8krttC5uOPUoiO40+8yeNThwLP1fh6eq2GV24fym1WJ/5zdSeHiHS2KZv8knwiPSMRQlBsLWZz8mZe2PIqkWYzL2Tn0NVUguGKDy4tkQZVRCMGqD/Do0eeWQzseiNEDlL9wFu/hiveVY9XFFVsIgdWFC+d4XyRBrWWyDVNz60GqDHMhxepC5RnZwjumKkuRHqGqklBK19VH9/0kxq9cTaFGRC/WU0gGvGS6na5UCMH70jV/WMrueQWEuEScn2kvvkmOXN/Qevjg1JSgt1oBEAXGIjf3XfhPXkyGrczvuXCEiuzNsXw1dpT2BV4cHg0/zcwisQcI/nFVvq38r3gjNliszN3WzzfbTxNQnYxAP7uTmg1kJZfQqSfK+E+rtw3tBVD2wY45N5Mx44Rc80kfO++i6Bnn3XImABmq50Ztz/DqN3Lyrd90v0GRjx1Dzf3jahwbOyO/Rx57yPCD+8g1d2fo0+9zUNThqCpoz/ZZleq9UUn5CewIWkDLdxb0CuoV3k8cxkphSn8cOgH5h+fj9luVoUaQUJBAjbFRpQNZmnC8Gt3pZomPumrS8PdURmKos6u215xpqnq/HvUn+1PHgQXb9XN8b+uMH4a9LuvUc2tN+Yi+LQrlBSo/vbhL6j+5k87q19aU+aqvuyUverffg/AuHcqjrHzB/j7iTNuoKr4tKsaU9/rTjWUsIlxybs+8hYtImfuL/j+3+0EPvMMitlM0fbtGMLDMbRqdV4W4dbTWTwydw+ZhSWM7hDE61d1JNxX9Y96uVx4xmpX7GiEBr1Ww/8NjOL2AZHkGi0k5GVyMG8dBWYj1pxBHE0pZk98Drd/v52uYV60C/Kge4Q37YM9Sc83sTMuh70JuTw7rh39S2fruUYzW09n0SfKFz/382eYzu3a4XX9dWTP/hHvSZNwatPGIa/bj7OXMXzPcoqGjqFH+BKS1ht4fO88Pp/ly7U9HuWXbXG09HOlvc5E6r334GezcXToVfR54XFGt2pR6bgWu4XDWYeJ9IjE29n7gsdUJdJmm5nP93zOT0d+wmq3AhDuEc6c8XPIM+ex9PRSTuedZk3CGgCujr6adj7t2JC0ARedC2Mix9DNpqHP0ldwve71S6anYbUIoc6kz2bgY3BgnhrnPO4diNukbq/vAlxTwOAG969XMzTXvQ/tJ6ihdXar2hZNq1fjuUHtCL5njrpdo4VDC+HgPIjboqbvl7mBqsIvWhVqOaNWccSMWrHZyPntN/Lmzcd0+DCuffrQ4rvvOJBWhK+rgSj/ipEZiqKwNyGXv/enMGtzLJF+rnx0Q7dqu1Pvz9jPm1veJL4gnj7BfZjafioDQwdispr44eAPzDw4kxJbCQBdA7ryaI9HCXENY/m+ElYdSed0RiGZheby8Qw6DZ7OOorNNt6+tjMHEvP5bUc8RWYbOo3g9gFRvHZVx/PssObkcPqK8Riio4mY9QOK0Uje4r/xGDMafbC6OGQrLMReUIA+JOSC95KXnMqez75DXHUt607n0uezlwmghB5zP0H78zjsVsHR7b3JT81h3tBbuWbjb9iEBkWvx8OYj8t3s+jUv9v5A2eeINPJnS3pO0kzpjHvyFySijMAtZPJ8LDhXNnqSlp5V/8BSC1K5em1T7M/cz+TWl7J3S5RxOQc59m0NQS7tyC5MBmbYiPELYShYUO5s9OdhGz8DI78rdZz7nGbWuT9t9vUsqFPH7vk/I215q9HYe9ceHAzbP4Mji6BZ0+Xl0i95MlLhE86wRXvq78Yds2CFxMqJs/EboJZE6D7VHUxsCBZDR2MGgJ97qlZzY4lz8COb+Gmn9WsyyZGk4/6KMsYtJvNFG3ejOnQIQr/XY3p8GGcu3TBY9RINncazsv/xlFgsqohn51DaBvkgU1RSM0rZsOJTFLyTOg0giu7hvDWpM7n1S/OKs7Cx9kHjdBQaC7ksz2f8evRXwlwDWBo2FA2Jm0ktSiVtj5tSSxIxGg1ckXUFdzd5W7i8+N5ZdMrFFtVd8hrA17jhrY3oCgKMZlFnMooIsTLmZb+rhzNiuGJnxJJyC5GrxVc0TmEyb3CWLQniT/3JPH1rb24ovP54VF5i/8m+dlncRs4EEtyMubYWNDrce3RA3txMaYjR8BqJfiN1/G68SaOpRUQHeCOQachJyuPndfeRFh6HHkGV0q0BvwsRlpM/xxf3V7Vz6c1UBI+maMfbcFgKaHIP5gC7wB8Tx8h67n/MPKO686zKTN2A4sW3My3Pj4UlU6SO5WUMDW/gDQndzZ7+bHbXohNCPr7dOD9MV/h5+J33jgAO1N38vS6pzFZTbw94HXGbPhKjfEFVnv58bSfJyMiRvJyv5fPjHF8Ocy9EVr0hPxkKExV60jbzOpCU997a/z/7JKlKFMtz+rio8Y5RwyAm+c2tlWO5dOuENJVfY91znDnORUqFEVNvEk/pIb+jX4dWo2s3ZdVWajiQ1ubZPutJifUxQcPkfvbrwiDEyUnT2LctQuthweKxYK9sBCEwBAVhf8jD+M89gqWHkzh6d/30SvShzsGtuRgch4/b40jv1S0/dwMdA/3ZnznEEZ3CMLLVRXodGM6m5I2cSjrEFtTthKXH4e3kzct3FtwOvc0JbYSbm5/M4/2eBR3gztmm5mfjvzEuoR1tPVpy9iosfQJ7lNud1ZxFidzTzLr0Cw2JW3ilf6vMLntZDRCg8VuYeaBmfxx7A/Si9O5o8P9dHKdzKDWfuVfGBabnau/2ER2UQkrnxqG5wUK4ef8/jupr7+B1suL4DfewLh9G6bDR9C4uuDcqRPGI8co3riBHdF9OKX1xO7tQ1s/Z3x2bCQqM46sux7Bb+0/aIsKiJr+BS6dO8Gca9UPgH8biN9GYrtPSV+6nK7v/Qedjw9GYwmurhVdMha7hTc2v8HfpxZjR2GYVcNDaUmEebfGs+O10Go4rHkbMk+SFd6Thek7+dpNR5hXFDMnzMHX2Zd/Yv7hq31fodfoyTfnk1qUSqRnJJ8O+5joVW+pC0lXfQZhvWHmOIzBHXHteK2alKB3ARR15d/NH+5bCwg1gy9hOwx6DEIuMPu/XDm+XM1ATDsME6ZBt5sa2yLHsuBBOL5MDdsb8LDaAedckveq4YZdb6rbWkRxjlqAqvfdTbLIVJMS6sL160l8/Al1IU+rRR8chNvAgdiN6kzVY8xoXPv0wajR89KfB1hyIAWbXaF3pA8/3t0XV8OZN8hmV1AUBZ224rfq7rTdTN87nR2pO1BQcNO70T2wO32C+hCTF0OqMZXW3q2Z2HIiXQKqWYC4ACariUdWP8K2lG1EeUbRL6QfR7KPsD9jP0PDhmKz29iSsoXvx31Pr6BeFFmKSCpMoq1PW/Yl5HLtl5uY1COUj2/sfsHxjbt3ow8ORt/ijJ/YZleYuy2OL1Yc4eZNv9I/8xiepgKE3Q5ArlcA+rvupff9t6HY7WCzIfR6sJjgg0h1ASWin9oS6va/Ku28bLVbyTHlMG3HNJbFLuM2s45r9AG0u/0fNaGibJHrXPJT2PZVDx4ODiLarz0fDfuIGxbfQIBrAJGekXjoPYjwjGBqh6l4bP5SzWI7u/bD3rmw8EH1sZOXujqPUKM4bvpJnW1JLl/KYsNBXURsP7Fx7WkELv5ioqJgSUs7r65z7oKFpLzyCrRqzb93vEj7jlE46bTEZhXRr6UvLbxd+PdoOsfWx/HPwVRiMou4fUAU3cK8aBNexJLYBZzOPU2oeyjXtrkWN70bZfWickw5rIpfxbKYZexI3UGgSyAPdX+IURGjiPaORiMc589z1jnzzehvWBm/krlH5vJP7D8A/HfofxnfcjxFliJuXHwj9yy/B18XX7KKs7ApNt4c+CbXtrmWx0a14dNVJ+gT5csVnYLxctFXiK5w7dmz/HFijpFFe5P5e38KR1LyGdDKj35ffUy/Vr6gKNjy8lAUBb3vmbBBodGU/yQ0nlrFYhcdGyynaV8cxCN6N7VR6zlCbbFZmHt0LjP2zyDfrPaFeKrD/3Hn0rfgivvU8SoTaQDPEPqFDeWjrAM8yhFuWHwDJpuJz0Z8RpRX1Jnjjq9Qi/90uxkGPHJme7eb1UUerzD1n6R5EXlWJ/iwPpUf10xpkBl1Vz8/5begYLyunUTAI4+gCwoi86uvyPz8C4xdenJv9A1kc/7PfoNWg1kxotMXEexXRGjkdpKLT2K2mTFa1XA8J60TJbYS3PRuBLoGotfo0QgNJ3JOqCFbnlFMaj2JWzrcgovuAvGUDYSiKBXC/RLyE5h3Yh6ZxZkEuwWzM3UnR7KP8NuVvxHhEcVtM7ex+VQWAN3DvZl9Z99yl00ZS/an8Pz8/RSWWOkc6sk9g1txTfcWNUvEURQyTq3i4bWPc0SvxdfJh+ySHF40RHJL/EF45gQmxYZBa8Cu2Hlo1UNsSdnCoNBBjAgbQSvvVvQ5tUX1bT++X02qqI7Di+D325k1/GE+ilvMvV3u5bGej1Wwiel91XjW+9ZeON5V0jxRFPiovVoX5In9jW1No3DRXR+dXFyUp4ZdT//E/Wg1Aqe2bTEdPMiJbkN4KnI8rVrH0qZlPHrFGwWwawoxFweQZcwlxrwCs12NsvB38Wdk+Ej0Wj3tfNrRO6g3oR6hHMw8yIKTCygwF2CxWbDYLbT1acsVLa+gnU+7JtlPLq0ojesXX09Lz5bMmTCHPKOFv/YlkWO08MXqk7QNdufnu/uXi/WivUk8/uteuod789mUHuenXxfnqsV8BjwMftHsz9jP78d+Z2/6XsZYNfgm7WWWpxsFWh3T+r3C4HaTeWLtE6xPWMdrGZm0HfEGDx+eQZCLP+1D+rDw5EJe7fcKN3p3UrPgirNh9lXg7A0PbKjZTVrN8FE7lIgBHB7zMu1926M9O+Ps1GrVX37tN2q2mURyNrtmq6F3PW5tbEsahYsu1AGBPkr0u9egiffkwQPp9Dh1lDn9OvBXVxu+Prnk21Lwd/GnyFIEgKfBkzRjGgLBxFYTGdhiIK56VwaHDsZJ2zDpx43Bz0d+5v3t7zN3wtwKvvE1R9O5b85OhrUN5Nvbe5GWX8LYT9bROtCdX+8bcMGOJ9sX3MmC5HW0dQ0hq/MkZh/+ETe9Gx21HuwsTkERgp4uwTw/4HU6hg8GwGgx8sTqx9iSug2dohBktWHSaMjSarjRqyOvntwDhWlqvQSdk1omc+q8mlVzK+PfN2HDx2rRm4Pz1JZHfe6F3nepRXiSdqkV03SXz/sqkTiCiy7Ubi3dlEHTBpFSlKJusGlAa6eFayStfMKZ0HICE1tNRCM05S6DbFM2ZpuZYLcqqnrVh6xTajxqXoIa/hM9Si0kHrdJ7T7RfapaXawBM9sKzYWM+mMUoyNH887gihlW32+M4c2/D3Nj7zAOp+RzKr2IpY8PoeVZ8eJWu5UVsSv47cBMducex02hPGTuJpcInsrOwzX1AAkdJ2Ic/TrtfM+vDmaz25jxxzXsyz7K261uQLf7R9a5uTEhOw199EjodB0k7lBX16/5ovZhTIUZ8GkXdfEvcQd4hatJBghAUUtkjniplq+cRHL5c9GFule4i7JrzWLSQrqwPGYN62L3cV/P6+nX4oI2NBw2qyrG5kL44041kqDNWCjKUON33YMhejicXA35iWqpyas+LW9+2hC8vfVtFpxYwKobVuHjfCYZR1EUHvxpN/8cSqWFlzMvTOjA1d3ORH0Umgt5et3TbE7eTDg6bsk3csOd60j//TaMqXtoZ9eqM9/IQWrXjqr8v6Z8tQNzaC/15+bix2DwUzDyVcckUSx7HrZ9rdZneGgLFKSp/ftSD8CVn5S3xJJIJGe4+OF5EW7Kzrt0aujVqNcrZhiBWmw8diP0f6jqSILiHEg/onap2PcLHFlcWtS8G1z1v6rrye77VY0uyIlVn7v4wP/9DcGlqaZWM2h0qjDZbWqc6pp3VGGf+LEaFeEZqu5b+aoaw3nTnKrr+NaAU7mnmLRoEmMix/D+kPcxaM/UGjZb7WQWlhDi5YwQAqPFyOzDs9mZupP4gngyjBm83O0Rrl/wDJrhL6i1EQrTIXGn+qvAyb1uRhVlnemE4Qjyk+H321XhryQMUCKRVOTiC3WvXsrON4er/smwvjDpS/VncOIOtQdb3Eb1QJ8otRiLR4hahOVsQc9Lgu/Hqa4KUCMF2oxTF7oO/K4WdBnwCAx56vzedGVFwkO6q4XV9c5qyM/ZpSIvhClfbYJZVk9B56IKfEEyaJ3UyIf/+/vCVcpqQVlz1SjPKMw2M538O/HWoLdKww1V9qbv5am1T5FRnEEX/y74OPswtcNUBsbvU7OrHt4BAW3rZYdEImk6NF7Cy8E/4a/HwFyg7nDxUWepna9X6zYsehTy4tV9LYepboeYDWru/6E/1dnixI/VIi2RA1Rhh7M6KP+m1uUd9aq6L2WfOoPe8Z3qb75hVu3b/FhL1Jq2halqFljmcVXsnT3hp8ng6qsmYNSktkAVLDm9hLlH5uLv4s+6RDUT8svRX+Lv4s+e9D08sPIB/F38eWfwO3QPPOtaM8eqX1IPbqrX9SUSSdOicTMTs06pbohWw8+voWs1q7PVU2tg6TOqIKtmqR0pJn9fdZWwxJ3q7DJxx5ltZQuFk793fLGepN3w261qp+YhT6uVzRxwjQ2JG3h63dOEuIVwS/tb+HDnhwS7BfPd2O8Icjtr9l5WvGbkqzD0mXpfVyKRNB2aVAp5pcRvVWfT7a5QSxbWNBZaUeDkKkCos3QXn4bN4y/MgKVPq8kdGr3qsw7pprpZbGZI3Q/Je1R3T+vRNR52R+oOHv73YYqtxfQM7MlHwz/C3+WsRTerWU3/PrYUHtt9SZZqlEgklXNpCPWlRswG9QsiP0md0ZctWvq0VAVbaNQqXbVY4NuXsY/dabu5teOt6DVn+esL02HRI3BiOYz/L/S737H3IpFIGh0p1BcDi0lN4hBC/XXw/Tg1yWP8f9WMvL0/q49rGzVycD789ThYi9V6vc2hrKdE0gy55Du8XBKc7auO6K8uQG6foS6omnLV7Ro9TJ5Z8zGN2fD3k2p50uu+Bf/WDjVZIpFcGkihbiiu+ABaj4F9cyGwE1iK1HrCve6AlkNqNsbGT9SQwWu+kCItkTRjpFA3FBoNtB2r/gMwG1U3xt9Pwv3r1H5xVZGboM7Iu90MQZ0a3l6JRNJkuUyarl0CGFzh6i8g6yQseVqNVqkMu10toi+0si6GRCKRQn1RaTVMTfve9wts+Khysd76JcRugPHvg3f4xbVRIpE0OaTr42Iz9Fk123H1W5B+WE2dL8pQa5p0uFqtb7LyVWh/pdpxWyKRNHukUF9sNFq47js1YWXDx6rfGtTCU/t/Ux9Hj4LrZjTJBpwSieTiI4W6MdBoYOQrMPQ5tbyqwUPNqNw3F7JjYPiLoDNUP45EImkWSKFuTHSGiqngPW9vPFskEkmTpUaLiUKIK4QQx4QQJ4UQLzS0URKJRCI5Q7VCLYTQAtOB8UBH4GYhRMeGNkwikUgkKjWZUfcFTiqKclpRFDPwK3BNw5olkUgkkjJqItShQMJZzxNLt1VACHGfEGKnEGJnRkaGo+yTSCSSZo/DEl4URZmhKEpvRVF6BwQEOGpYiUQiafbURKiTgLPT48JKt0kkEonkIlATod4BtBFCtBRCGIApwF8Na5ZEIpFIyqg2jlpRFKsQ4hFgOaAFvlcU5VCDWyaRSCQSoIE6vAghCoBjDhrOC8hz0FgA/kCmg8aSttUNaVvdkLbVjUvFtkhFUS68wKcoisP/ATsdONYMaZu0TdombWvOtl0KZU4XN7YBVSBtqxvStrohbasbl7xtTV6oFUVpsi+ytK1uSNvqhrStblwOtjWUUM9ooHEdgbStbkjb6oa0rW5I286iQRYTJRKJROI4mrzrQyKRSJo7UqglEomkiVPTetThQog1QojDQohDQojHS7f7CiFWCiFOlP71Kd0uhBCfldav3i+E6HnWWDYhxN7Sf/XOcHSUbUKIEWfZtVcIYRJCTGoKtpXu+0AIcbD03031sauOtrUXQmwRQpQIIZ45Z6zvhRDpQoiD9bXLkbYJIZyFENuFEPtKx/lPU7GtdF+sEOJA6f+3nU3FNiFEu3M+C/lCiCeagm2l+x4v/Rwcqq9ddbRtaunn84AQYrMQottZYzn0s1BODWP9QoCepY89gOOotan/C7xQuv0F4IPSxxOAZYAA+gPbzhqr0MExjQ6z7awxfYFswLUp2AZMBFaiZpK6oab1e15k2wKBPsA7wDPnjDUU6AkcbKT39IK2lb6O7qWP9cA2oH9TsK10Xyzg34ifhUptO2tMLZCKmozR6LYBnYGDgGvp52EV0Poi2zYQ8Cl9PJ6K+ubQz0L5uHW8sUXAGNTsw5CzbvZY6eNvgJvPOv7s4xwq1I607axt9wE/NxXbgGeBV8/aPhO48WLadtZxb1zoQw1EOfo/p6NsK93nCuwG+jUV23CwUDfQ6zYW2NRUbANuAGae9fxV4LnGsK10uw+QdM42h38Wau2jFkJEAT1QZydBiqKklO5KBYJKH1dVw9pZqHWrt4p6uhYawLYypgC/NCHb9gFXCCFchRD+wAgqVjS8GLY1CvW1TQihFULsBdKBlYqibGsqtgEKsEIIsUsIcZ+j7HKQbWU01mehMg4CQ4QQfkIIV9RfoY35Wbgb9Vdwg1Kr5rZCCHdgPvCEoij5QojyfYqiKEKImsT6RSqKkiSEaAWsFkIcUBTlVK2sbjjbEEKEAF1Qi1A5hPrapijKCiFEH2AzkAFsAWxNwbaGxBG2KYpiA7oLIbyBBUKIzoqi1Nt/6KDXbXDpZyEQWCmEOKooyvomYhtCrZZ5NfBifW1ylG2KohwRQnwArACKgL000mdBCDECVagHO+L6VVHjGbUQQo96Ez8rivJn6ea0UmErE7j00u2V1rBWFKXs72lgLeq3V71wlG2l3AgsUBTFUl+7HGmboijvKIrSXVGUMai+1+MX2baLiqNtUxQlF1gDXNFUbDvrs5AOLEBte9ckbCtlPLBbUZS0+trlSNsURZmpKEovRVGGAjk0wmdBCNEV+A64RlGUrPpevzpqGvUhUP2iRxRF+fisXX8B/1f6+P9QfTtl228XKv2BPEVRUoQQPkIIp9Ix/YFBwOH63ICjbDvrvJtx0E89B75uWiGEX+mYXYGuqDOKi2nbRcNRtgkhAkpn0gghXFD9jkebiG1uQgiPsseovuB6zfQb4D1tzM9CVWMFlv6NAK4D5l5M20qv+ydwm6Io9f6SqBE1dK4PRvWn7Uf9qbEX1TfkB/wLnEBdffUtPV6gdi4/BRwAepduH1j6fF/p37vr62R3lG3KmUWAJEBTX7sc/Lo5o36hHQa2At0bwbZgVJ95PpBb+tizdN8vQApgKd1er/fVUbahfqHtKR3nIPBaU3ndgFaon4N9wCHg5aZiW+k+NyAL8Gqkz0JVtm1A/SzsA0Y1gm3foc7ky47dedZYDv0slP2TKeQSiUTSxJGZiRKJRNLEkUItkUgkTRwp1BKJRNLEkUItkUgkTRwp1BKJRNLEkUItuewQQrwhzqm4ds7+SUKIjhfTJomkPkihljRHJqFWR5NILglkHLXkskAI8TJq9lg6amGrXUAeaiVEA3ASuA3oDvxdui8PuL50iOlAAGAE7lUUpV4ZjBKJI5FCLbnkEUL0AmYB/VALje0GvgZ+UErrMAgh3gbSFEX5XAgxC/hbUZR5pfv+BR5QFOWEEKIf8J6iKCMv/p1IJBemVtXzJJImyhDUQlpGAHGmc1DnUoH2Bty5QEXE0oppA4E/zqqW5tTQBksktUEKteRyZhYwSVGUfUKIO4DhFzhGA+QqitL94pklkdQOuZgouRxYD0wSQriUVqS7qnS7B5BSWsJy6lnHF5TuQ1GUfCBGCHEDlPet7HbxTJdIqkcKteSSR1GU3cBvqNXUlqH2lAS1TdM2YBMVy5v+CjwrhNgjhIhGFfG7hRBlleyuuVi2SyQ1QS4mSiQSSRNHzqglEomkiSOFWiKRSJo4UqglEomkiSOFWiKRSJo4UqglEomkiSOFWiKRSJo4UqglEomkifP/jmtokb5CYekAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "(1+rets_risk_parity[[\"best\", \"worst\", \"mkt\", \"ls\"]]).cumprod().plot()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Excess returns and Sharpe ratios"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 188,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "best     0.7764\n",
       "worst    0.3780\n",
       "mkt      0.6513\n",
       "ls       0.8078\n",
       "dtype: float64"
      ]
     },
     "execution_count": 188,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "xrets = rets[[\"best\", \"worst\", \"mkt\", \"ls\"]].subtract(rets.rf, axis=0)\n",
    "sharpes = np.sqrt(12)*xrets.mean()/xrets.std()\n",
    "sharpes.round(4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 189,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "best     0.8294\n",
       "worst    0.4152\n",
       "mkt      0.6513\n",
       "ls       0.8076\n",
       "dtype: float64"
      ]
     },
     "execution_count": 189,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "xrets = rets_risk_parity[[\"best\", \"worst\", \"mkt\", \"ls\"]].subtract(rets_risk_parity.rf, axis=0)\n",
    "sharpes = np.sqrt(12)*xrets.mean()/xrets.std()\n",
    "sharpes.round(4)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "130/30 drawdowns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 191,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 192,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, ' % Drawdown')"
      ]
     },
     "execution_count": 192,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ls = (1+rets.ls).cumprod()\n",
    "lsmax = ls.expanding().max()\n",
    "drawdown = ls/lsmax - 1\n",
    "\n",
    "fig, ax1 = plt.subplots()\n",
    "\n",
    "ax2 = ax1.twinx()\n",
    "ax2.set_yscale(\"log\")\n",
    "ax2.plot(rets.index.to_timestamp(), ls, 'b-')\n",
    "ax1.plot(rets.index.to_timestamp(), drawdown, 'r-')\n",
    "\n",
    "ax1.set_xlabel('Date')\n",
    "ax2.set_ylabel('Accumulation (log scale)', color='b')\n",
    "ax1.set_ylabel(' % Drawdown', color='r')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 193,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, ' % Drawdown')"
      ]
     },
     "execution_count": 193,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ls = (1+rets_risk_parity.ls).cumprod()\n",
    "lsmax = ls.expanding().max()\n",
    "drawdown = ls/lsmax - 1\n",
    "\n",
    "fig, ax1 = plt.subplots()\n",
    "\n",
    "ax2 = ax1.twinx()\n",
    "ax2.set_yscale(\"log\")\n",
    "ax2.plot(rets_risk_parity.index.to_timestamp(), ls, 'b-')\n",
    "ax1.plot(rets_risk_parity.index.to_timestamp(), drawdown, 'r-')\n",
    "\n",
    "ax1.set_xlabel('Date')\n",
    "ax2.set_ylabel('Accumulation (log scale)', color='b')\n",
    "ax1.set_ylabel(' % Drawdown', color='r')"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.6"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}