fama_macbeth.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "e9bc2988",
   "metadata": {},
   "source": [
    "# Fama-Macbeth Cross-sectional Regressions\n",
    "\n",
    "_UNDER CONSTRUCTION_\n",
    "\n",
    "\n",
    "- Beta, tests of the CAPM\n",
    "- Polynomial regression, feature transformations\n",
    "- Risk premiums from cross-sectional regressions\n",
    "- Kernel regression and LOOCV\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "2c3dc256",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "from scipy.stats import skew, kurtosis\n",
    "import pandas as pd\n",
    "from pandas import DataFrame, Series\n",
    "import matplotlib.pyplot as plt\n",
    "import statsmodels.formula.api as smf\n",
    "from pandas_datareader.famafrench import FamaFrenchReader\n",
    "from tqdm import tqdm\n",
    "from finds.database import SQL, RedisDB\n",
    "from finds.structured import Signals, Benchmarks, CRSP\n",
    "from finds.busday import BusDay\n",
    "from finds.backtesting import RiskPremium\n",
    "from finds.filters import winsorize\n",
    "from finds.econs import least_squares\n",
    "from finds.misc import Show\n",
    "from secret import credentials, paths, CRSP_DATE\n",
    "show = Show(ndigits=4, latex=None)\n",
    "VERBOSE = 0\n",
    "#%matplotlib qt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "5847f30e",
   "metadata": {
    "lines_to_next_cell": 2
   },
   "outputs": [],
   "source": [
    "# open connections\n",
    "sql = SQL(**credentials['sql'], verbose=VERBOSE)\n",
    "user = SQL(**credentials['user'], verbose=VERBOSE)\n",
    "rdb = RedisDB(**credentials['redis'])\n",
    "bd = BusDay(sql, verbose=VERBOSE)\n",
    "crsp = CRSP(sql, bd, rdb=rdb, verbose=VERBOSE)\n",
    "signals = Signals(user, verbose=VERBOSE)\n",
    "bench = Benchmarks(sql, bd, verbose=VERBOSE)\n",
    "imgdir = paths['images']\n",
    "LAST_DATE = bd.endmo(CRSP_DATE, -1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "487c89ba",
   "metadata": {},
   "outputs": [],
   "source": [
    "import warnings\n",
    "warnings.simplefilter(action='ignore', category=FutureWarning)\n",
    "## Retrieve market and test asset returns\n",
    "asset_names = ['25_Portfolios_ME_BETA_5x5']\n",
    "# '25_Portfolios_ME_VAR_5x5', '25_Portfolios_ME_RESVAR_5x5', '25_Portfolios_5x5']\n",
    "\n",
    "test_assets = {asset: FamaFrenchReader(asset, start=1900, end=2099).read()\n",
    "               for asset in asset_names}\n",
    "#for a in test_assets.values():\n",
    "#    for p in range(2):\n",
    "#        a[p].index = a[p].index.strftime('%Y%m').astype(int)\n",
    "\n",
    "mkt = FamaFrenchReader('F-F_Research_Data_Factors', start=1900, end=2099).read()\n",
    "mkt = mkt[0].rename(columns={'Mkt-RF': 'BETA'})\n",
    "#mkt.index = mkt.index.strftime('%Y%m').astype(int)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "1622560d",
   "metadata": {
    "lines_to_next_cell": 2
   },
   "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>of months</th>\n",
       "      <th>annual mean</th>\n",
       "      <th>annual stdev</th>\n",
       "      <th>sharpe ratio</th>\n",
       "      <th>skewness</th>\n",
       "      <th>excess kurtosis</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Summary statistics of Mkt-RF monthly returns</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>1926-2023</th>\n",
       "      <td>1164</td>\n",
       "      <td>8.1160</td>\n",
       "      <td>18.5122</td>\n",
       "      <td>0.4384</td>\n",
       "      <td>0.1572</td>\n",
       "      <td>7.4086</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1926-1949</th>\n",
       "      <td>282</td>\n",
       "      <td>8.6630</td>\n",
       "      <td>26.6461</td>\n",
       "      <td>0.3251</td>\n",
       "      <td>0.4932</td>\n",
       "      <td>5.5027</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1950-1974</th>\n",
       "      <td>300</td>\n",
       "      <td>6.4956</td>\n",
       "      <td>13.5968</td>\n",
       "      <td>0.4777</td>\n",
       "      <td>-0.2875</td>\n",
       "      <td>0.8862</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1975-1999</th>\n",
       "      <td>300</td>\n",
       "      <td>10.7464</td>\n",
       "      <td>15.3718</td>\n",
       "      <td>0.6991</td>\n",
       "      <td>-0.7028</td>\n",
       "      <td>3.2154</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2000-2023</th>\n",
       "      <td>282</td>\n",
       "      <td>6.4945</td>\n",
       "      <td>15.9814</td>\n",
       "      <td>0.4064</td>\n",
       "      <td>-0.4912</td>\n",
       "      <td>0.7481</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                              of months  annual mean   \n",
       "Summary statistics of Mkt-RF monthly returns                           \n",
       "1926-2023                                          1164       8.1160  \\\n",
       "1926-1949                                           282       8.6630   \n",
       "1950-1974                                           300       6.4956   \n",
       "1975-1999                                           300      10.7464   \n",
       "2000-2023                                           282       6.4945   \n",
       "\n",
       "                                              annual stdev  sharpe ratio   \n",
       "Summary statistics of Mkt-RF monthly returns                               \n",
       "1926-2023                                          18.5122        0.4384  \\\n",
       "1926-1949                                          26.6461        0.3251   \n",
       "1950-1974                                          13.5968        0.4777   \n",
       "1975-1999                                          15.3718        0.6991   \n",
       "2000-2023                                          15.9814        0.4064   \n",
       "\n",
       "                                              skewness  excess kurtosis  \n",
       "Summary statistics of Mkt-RF monthly returns                             \n",
       "1926-2023                                       0.1572           7.4086  \n",
       "1926-1949                                       0.4932           5.5027  \n",
       "1950-1974                                      -0.2875           0.8862  \n",
       "1975-1999                                      -0.7028           3.2154  \n",
       "2000-2023                                      -0.4912           0.7481  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Show summary statistics of Mkt-Rf, every 25 years\n",
    "periods=[(1925, 2024), (1925, 1949), (1950, 1974), (1975, 1999), (2000, 2024)]\n",
    "df = []\n",
    "for (beg, end) in periods:\n",
    "    data = mkt[(mkt.index >= f\"{beg}-01\") & (mkt.index <= f\"{end}-12\")]['BETA']\n",
    "    df.append(DataFrame({'of months': len(data),\n",
    "                         'annual mean': 12*np.mean(data),\n",
    "                         'annual stdev': np.std(data)*np.sqrt(12),\n",
    "                         'sharpe ratio': np.mean(data)*np.sqrt(12)/np.std(data),\n",
    "                         'skewness': skew(data),\n",
    "                         'excess kurtosis': kurtosis(data, fisher=True)},\n",
    "                        index=[f\"{str(data.index[0])[:4]}-\" + \n",
    "                               f\"{str(data.index[-1])[:4]}\"]))\n",
    "show(pd.concat(df),\n",
    "     caption=\"Summary statistics of Mkt-RF monthly returns\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "69ee7ba5",
   "metadata": {},
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "da380444",
   "metadata": {},
   "source": [
    "## Cross-sectional regressions\n",
    "\n",
    "### Feature transformations\n",
    "\n",
    "A simple way to directly extend the linear model to accommodate\n",
    "non-linear relationships, using polynomial regression, is to include\n",
    "transformed versions of the predictors in the model, such asa\n",
    "quadratic term or several polynomial functions of the predictors, \n",
    "and use standard linear regression to estimate coefficients in order to produce a non-linear fit.  The CAPM predicts that these coefficients are zero.\n",
    "\n",
    "Raw polynomial terms may be highly correlated with each other: _orthogonal polynomials_ transform the raw data matrix of polynomial terms to another whose columns are a basis of orthogonal terms which span the same column space. For example, regress the second predictor on the first and replace its column with the residuals, then regress the third predictor on the first two and replace its column with the residuals, and so on.\n",
    "\n",
    "Other feature transformation approaches include:\n",
    "- dummy or binary indicator variable\n",
    "- categorical variables with two or more levels\n",
    "- binarization or turning a categorical variable into several binary variables (4)\n",
    "-  Legendre polynomals which are defined as a system of orthogonal polynomials over the interval $[-1, 1]$ \n",
    "-  interaction term constructed by computing the product of the values of the two variables to capture the effect that response of one predictor is dependent on the value of another predictor."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "2f6e5914",
   "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 tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr:last-of-type th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th colspan=\"4\" halign=\"left\">Value-weighted Test Assets</th>\n",
       "      <th colspan=\"4\" halign=\"left\">Equal-weighted Test Assets</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>Intercept</th>\n",
       "      <th>BETA</th>\n",
       "      <th>BETA2</th>\n",
       "      <th>RES</th>\n",
       "      <th>Intercept</th>\n",
       "      <th>BETA</th>\n",
       "      <th>BETA2</th>\n",
       "      <th>RES</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Fama-Macbeth Monthly CSR 1926-07 to 2023-06</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=\"3\" valign=\"top\">25_Portfolios_ME_BETA_5x5</th>\n",
       "      <th>mean</th>\n",
       "      <td>0.6599</td>\n",
       "      <td>0.0861</td>\n",
       "      <td>-0.8213</td>\n",
       "      <td>0.1061</td>\n",
       "      <td>0.8044</td>\n",
       "      <td>-0.0080</td>\n",
       "      <td>-0.8349</td>\n",
       "      <td>0.1315</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>stderr</th>\n",
       "      <td>0.1637</td>\n",
       "      <td>0.2374</td>\n",
       "      <td>0.2630</td>\n",
       "      <td>0.0675</td>\n",
       "      <td>0.1561</td>\n",
       "      <td>0.2295</td>\n",
       "      <td>0.2597</td>\n",
       "      <td>0.0615</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>tstat</th>\n",
       "      <td>4.0310</td>\n",
       "      <td>0.3625</td>\n",
       "      <td>-3.1230</td>\n",
       "      <td>1.5724</td>\n",
       "      <td>5.1518</td>\n",
       "      <td>-0.0349</td>\n",
       "      <td>-3.2146</td>\n",
       "      <td>2.1390</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                   Value-weighted Test Assets   \n",
       "                                                                    Intercept   \n",
       "Fama-Macbeth Monthly CSR 1926-07 to 2023-06                                     \n",
       "25_Portfolios_ME_BETA_5x5                   mean                       0.6599  \\\n",
       "                                            stderr                     0.1637   \n",
       "                                            tstat                      4.0310   \n",
       "\n",
       "                                                                             \n",
       "                                                      BETA   BETA2     RES   \n",
       "Fama-Macbeth Monthly CSR 1926-07 to 2023-06                                  \n",
       "25_Portfolios_ME_BETA_5x5                   mean    0.0861 -0.8213  0.1061  \\\n",
       "                                            stderr  0.2374  0.2630  0.0675   \n",
       "                                            tstat   0.3625 -3.1230  1.5724   \n",
       "\n",
       "                                                   Equal-weighted Test Assets   \n",
       "                                                                    Intercept   \n",
       "Fama-Macbeth Monthly CSR 1926-07 to 2023-06                                     \n",
       "25_Portfolios_ME_BETA_5x5                   mean                       0.8044  \\\n",
       "                                            stderr                     0.1561   \n",
       "                                            tstat                      5.1518   \n",
       "\n",
       "                                                                            \n",
       "                                                      BETA   BETA2     RES  \n",
       "Fama-Macbeth Monthly CSR 1926-07 to 2023-06                                 \n",
       "25_Portfolios_ME_BETA_5x5                   mean   -0.0080 -0.8349  0.1315  \n",
       "                                            stderr  0.2295  0.2597  0.0615  \n",
       "                                            tstat  -0.0349 -3.2146  2.1390  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "periods = [mkt.index[0], '1963-07']\n",
    "out = {period: [] for period in periods}\n",
    "for period in periods:\n",
    "    \n",
    "    for asset in asset_names:\n",
    "        by_asset = []\n",
    "    \n",
    "        for p, wt in enumerate(['Value-weighted Test Assets', 'Equal-weighted Test Assets']):\n",
    "            # subtract riskfree, and stack data as thin dataframe\n",
    "            f = test_assets[asset][p]\n",
    "            df = f.sub(mkt['RF'], axis=0).dropna().copy()\n",
    "            rets = df.stack().reset_index(name='ret')\\\n",
    "                             .rename(columns={'level_1':'port',\n",
    "                                              'level_0':'Date'})\n",
    "            data = df.join(mkt[['BETA']], how='left')\n",
    "            data = data[data.index >= period]\n",
    "\n",
    "            # estimate test assets market betas from time-series of returns\n",
    "            betas = least_squares(data,\n",
    "                                  y=df.columns,\n",
    "                                  x=['BETA'],\n",
    "                                  stdres=True)[['BETA', 'stdres']]  \n",
    "        \n",
    "            # orthogonalize beta^2 and residual-volatility regressors\n",
    "            betas['BETA2'] = smf.ols(\"I(BETA**2) ~ BETA\",\n",
    "                                     data=betas).fit().resid\n",
    "            betas['RES'] = smf.ols(\"stdres ~ BETA + BETA2\", data=betas).fit().resid\n",
    "            r = rets.join(betas, on='port')\\\n",
    "                    .sort_values(['port', 'Date'], ignore_index=True)\n",
    "        \n",
    "            # run monthly Fama MacBeth cross-sectional regressions\n",
    "            fm = r.groupby(by='Date').apply(least_squares,\n",
    "                                            y=['ret'],\n",
    "                                            x=['BETA', 'BETA2', 'RES'])\n",
    "\n",
    "            # time-series means and standard errors of the FM coefficients\n",
    "            sub = DataFrame({'mean': fm.mean(),\n",
    "                             'stderr': fm.sem(),\n",
    "                             'tstat': fm.mean() / fm.sem()}).T\n",
    "            sub.columns = pd.MultiIndex.from_tuples([(wt, col)\n",
    "                                                     for col in sub.columns])\n",
    "            sub.index = pd.MultiIndex.from_tuples([(asset, row)\n",
    "                                                   for row in sub.index])\n",
    "            by_asset.append(sub)\n",
    "        out[period].append(pd.concat(by_asset, axis=1))\n",
    "p = periods[0]\n",
    "show(pd.concat(out[p]),\n",
    "     caption=[f\"Fama-Macbeth Monthly CSR {p} to {df.index[-1]}\", None])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "fd32cad5",
   "metadata": {
    "lines_to_next_cell": 2
   },
   "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 tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr:last-of-type th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th colspan=\"4\" halign=\"left\">Value-weighted Test Assets</th>\n",
       "      <th colspan=\"4\" halign=\"left\">Equal-weighted Test Assets</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>Intercept</th>\n",
       "      <th>BETA</th>\n",
       "      <th>BETA2</th>\n",
       "      <th>RES</th>\n",
       "      <th>Intercept</th>\n",
       "      <th>BETA</th>\n",
       "      <th>BETA2</th>\n",
       "      <th>RES</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Fama-Macbeth Monthly CSR 1963-07 to 2023-06</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=\"3\" valign=\"top\">25_Portfolios_ME_BETA_5x5</th>\n",
       "      <th>mean</th>\n",
       "      <td>0.6599</td>\n",
       "      <td>0.0861</td>\n",
       "      <td>-0.8213</td>\n",
       "      <td>0.1061</td>\n",
       "      <td>0.8044</td>\n",
       "      <td>-0.0080</td>\n",
       "      <td>-0.8349</td>\n",
       "      <td>0.1315</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>stderr</th>\n",
       "      <td>0.1637</td>\n",
       "      <td>0.2374</td>\n",
       "      <td>0.2630</td>\n",
       "      <td>0.0675</td>\n",
       "      <td>0.1561</td>\n",
       "      <td>0.2295</td>\n",
       "      <td>0.2597</td>\n",
       "      <td>0.0615</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>tstat</th>\n",
       "      <td>4.0310</td>\n",
       "      <td>0.3625</td>\n",
       "      <td>-3.1230</td>\n",
       "      <td>1.5724</td>\n",
       "      <td>5.1518</td>\n",
       "      <td>-0.0349</td>\n",
       "      <td>-3.2146</td>\n",
       "      <td>2.1390</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                   Value-weighted Test Assets   \n",
       "                                                                    Intercept   \n",
       "Fama-Macbeth Monthly CSR 1963-07 to 2023-06                                     \n",
       "25_Portfolios_ME_BETA_5x5                   mean                       0.6599  \\\n",
       "                                            stderr                     0.1637   \n",
       "                                            tstat                      4.0310   \n",
       "\n",
       "                                                                             \n",
       "                                                      BETA   BETA2     RES   \n",
       "Fama-Macbeth Monthly CSR 1963-07 to 2023-06                                  \n",
       "25_Portfolios_ME_BETA_5x5                   mean    0.0861 -0.8213  0.1061  \\\n",
       "                                            stderr  0.2374  0.2630  0.0675   \n",
       "                                            tstat   0.3625 -3.1230  1.5724   \n",
       "\n",
       "                                                   Equal-weighted Test Assets   \n",
       "                                                                    Intercept   \n",
       "Fama-Macbeth Monthly CSR 1963-07 to 2023-06                                     \n",
       "25_Portfolios_ME_BETA_5x5                   mean                       0.8044  \\\n",
       "                                            stderr                     0.1561   \n",
       "                                            tstat                      5.1518   \n",
       "\n",
       "                                                                            \n",
       "                                                      BETA   BETA2     RES  \n",
       "Fama-Macbeth Monthly CSR 1963-07 to 2023-06                                 \n",
       "25_Portfolios_ME_BETA_5x5                   mean   -0.0080 -0.8349  0.1315  \n",
       "                                            stderr  0.2295  0.2597  0.0615  \n",
       "                                            tstat  -0.0349 -3.2146  2.1390  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "### Post-1963 time-period\n",
    "p = periods[1]\n",
    "show(pd.concat(out[p]),\n",
    "     caption=[f\"Fama-Macbeth Monthly CSR {p} to {df.index[-1]}\", None])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6d28f82b",
   "metadata": {},
   "source": [
    "__Clustered standard errors__"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "648f7ab9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                            OLS Regression Results                            \n",
      "==============================================================================\n",
      "Dep. Variable:                    ret   R-squared:                       0.000\n",
      "Model:                            OLS   Adj. R-squared:                  0.000\n",
      "Method:                 Least Squares   F-statistic:                     2.811\n",
      "Date:                Thu, 31 Aug 2023   Prob (F-statistic):             0.0379\n",
      "Time:                        10:40:36   Log-Likelihood:                -57761.\n",
      "No. Observations:               18000   AIC:                         1.155e+05\n",
      "Df Residuals:                   17996   BIC:                         1.156e+05\n",
      "Df Model:                           3                                         \n",
      "Covariance Type:            nonrobust                                         \n",
      "==============================================================================\n",
      "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
      "------------------------------------------------------------------------------\n",
      "Intercept      0.8044      0.189      4.256      0.000       0.434       1.175\n",
      "BETA          -0.0080      0.165     -0.048      0.961      -0.332       0.316\n",
      "BETA2         -0.8349      0.616     -1.356      0.175      -2.041       0.372\n",
      "RES            0.1315      0.051      2.567      0.010       0.031       0.232\n",
      "==============================================================================\n",
      "Omnibus:                     1559.773   Durbin-Watson:                   1.767\n",
      "Prob(Omnibus):                  0.000   Jarque-Bera (JB):            10364.480\n",
      "Skew:                          -0.063   Prob(JB):                         0.00\n",
      "Kurtosis:                       6.715   Cond. No.                         20.8\n",
      "==============================================================================\n",
      "\n",
      "Notes:\n",
      "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n"
     ]
    }
   ],
   "source": [
    "### Compare to robust cov\n",
    "ls = smf.ols(\"ret ~ BETA + BETA2 + RES\", data=r).fit()\n",
    "print(ls.summary())\n",
    "# print(ls.get_robustcov_results('HC0').summary())\n",
    "# print(ls.get_robustcov_results('HAC', maxlags=6).summary())\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "d8dbde79",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                            OLS Regression Results                            \n",
      "==============================================================================\n",
      "Dep. Variable:                    ret   R-squared:                       0.000\n",
      "Model:                            OLS   Adj. R-squared:                  0.000\n",
      "Method:                 Least Squares   F-statistic:                     1.696\n",
      "Date:                Thu, 31 Aug 2023   Prob (F-statistic):              0.195\n",
      "Time:                        10:40:36   Log-Likelihood:                -57761.\n",
      "No. Observations:               18000   AIC:                         1.155e+05\n",
      "Df Residuals:                   17996   BIC:                         1.156e+05\n",
      "Df Model:                           3                                         \n",
      "Covariance Type:            hac-panel                                         \n",
      "==============================================================================\n",
      "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
      "------------------------------------------------------------------------------\n",
      "Intercept      0.8044      0.199      4.045      0.000       0.394       1.215\n",
      "BETA          -0.0080      0.191     -0.042      0.967      -0.403       0.387\n",
      "BETA2         -0.8349      0.651     -1.282      0.212      -2.179       0.510\n",
      "RES            0.1315      0.067      1.952      0.063      -0.008       0.271\n",
      "==============================================================================\n",
      "Omnibus:                     1559.773   Durbin-Watson:                   1.767\n",
      "Prob(Omnibus):                  0.000   Jarque-Bera (JB):            10364.480\n",
      "Skew:                          -0.063   Prob(JB):                         0.00\n",
      "Kurtosis:                       6.715   Cond. No.                         20.8\n",
      "==============================================================================\n",
      "\n",
      "Notes:\n",
      "[1] Standard Errors are robust to cluster correlation (HAC-Panel)\n"
     ]
    }
   ],
   "source": [
    "print(ls.get_robustcov_results('hac-panel',\n",
    "                               groups=r['port'],\n",
    "                               maxlags=6).summary())\n",
    "# print(ls.get_robustcov_results('cluster', groups=r['port']).summary())\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fe4a8d01",
   "metadata": {},
   "source": [
    "## Kernel Regression\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0007b085",
   "metadata": {},
   "source": [
    "### LOOCV\n",
    "\n",
    "- Leave-one-out cross-validation"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c1fa0203",
   "metadata": {},
   "source": [
    "## Cross-sectional regressions and portfolio sorts\n",
    "\n",
    "Cross-sectional regression of monthly stock returns on fundamental stock exposures (winsored at 5% tail):\n",
    "- size: -log of market cap, standardized\n",
    "- value: book-to-market ratio, standardized \n",
    "- momentum: 12-month skip past month momentum, standardized\n",
    "- reversal: 1-month reversal, standardized\n",
    "\n",
    "compared to time series of portfolio-sort monthly returns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "967819b1",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  0%|          | 0/702 [00:00<?, ?it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 702/702 [16:00<00:00,  1.37s/it]\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 8 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "rebalbeg=19640601\n",
    "rebalend=LAST_DATE\n",
    "rebaldates = crsp.bd.date_range(rebalbeg, rebalend, 'endmo')\n",
    "loadings = dict()\n",
    "for pordate in tqdm(rebaldates):           # retrieve signal values every month\n",
    "    date = bd.june_universe(pordate)\n",
    "    univ = crsp.get_universe(date)\n",
    "    cap = np.sqrt(crsp.get_cap(date))\n",
    "    smb = -np.log(cap).rename('size')\n",
    "    hml = signals('hml', date, bd.endmo(date, -12))['hml'].rename('value')\n",
    "    #beta = signals('beta', pordate, bd.begmo(pordate))['beta']*2/3 + 1/3 #shrink\n",
    "    mom = signals('mom', pordate)['mom'].rename('momentum')\n",
    "    strev = -signals('strev', pordate)['strev'].rename('reversal')\n",
    "    df = pd.concat((strev, hml, smb, mom),  # inner join of signals with univ\n",
    "                   join='inner',\n",
    "                   axis=1).reindex(univ.index).dropna()\n",
    "    loadings[pordate] = winsorize(df, quantiles=[0.05, 0.95])\n",
    "\n",
    "# Compute coefficients from FM cross-sectional regressions\n",
    "riskpremium = RiskPremium(user, bench, 'RF', LAST_DATE)\n",
    "out = riskpremium(stocks=crsp,        # FM regressions on standardized scores\n",
    "                  loadings=loadings,\n",
    "                  standardize=['value' ,'size', 'momentum', 'reversal'])\n",
    "     \n",
    "# Compare time series of estimated risk premiums to portfolio-sort benchmark eturns\n",
    "benchnames = {'momentum': 'Mom(mo)',\n",
    "              'reversal': 'ST_Rev(mo)',\n",
    "              'size':'SMB(mo)',\n",
    "              'value': 'HML(mo)'}\n",
    "out = riskpremium.fit(benchnames.values())  # to compare portfolio-sorts\n",
    "riskpremium.plot(benchnames)\n",
    "plt.savefig(imgdir / 'fm.jpg')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "ce002f87",
   "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>Factor Returns</th>\n",
       "      <th>mean</th>\n",
       "      <th>stderr</th>\n",
       "      <th>std</th>\n",
       "      <th>count</th>\n",
       "      <th>tvalue</th>\n",
       "      <th>sharpe</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Fama-MacBeth Cross-sectional Regression Risk Premiums</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>reversal</th>\n",
       "      <td>0.0063</td>\n",
       "      <td>0.0005</td>\n",
       "      <td>0.0143</td>\n",
       "      <td>701</td>\n",
       "      <td>11.7877</td>\n",
       "      <td>1.5423</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>value</th>\n",
       "      <td>0.0022</td>\n",
       "      <td>0.0004</td>\n",
       "      <td>0.0112</td>\n",
       "      <td>701</td>\n",
       "      <td>5.1258</td>\n",
       "      <td>0.6706</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>size</th>\n",
       "      <td>0.0016</td>\n",
       "      <td>0.0007</td>\n",
       "      <td>0.0175</td>\n",
       "      <td>701</td>\n",
       "      <td>2.3589</td>\n",
       "      <td>0.3086</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>momentum</th>\n",
       "      <td>0.0019</td>\n",
       "      <td>0.0007</td>\n",
       "      <td>0.0175</td>\n",
       "      <td>701</td>\n",
       "      <td>2.8051</td>\n",
       "      <td>0.3670</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Factor Returns                                        mean  stderr     std   \n",
       "Fama-MacBeth Cross-sectional Regression Risk Pr...                           \n",
       "reversal                                            0.0063  0.0005  0.0143  \\\n",
       "value                                               0.0022  0.0004  0.0112   \n",
       "size                                                0.0016  0.0007  0.0175   \n",
       "momentum                                            0.0019  0.0007  0.0175   \n",
       "\n",
       "Factor Returns                                      count   tvalue  sharpe  \n",
       "Fama-MacBeth Cross-sectional Regression Risk Pr...                          \n",
       "reversal                                              701  11.7877  1.5423  \n",
       "value                                                 701   5.1258  0.6706  \n",
       "size                                                  701   2.3589  0.3086  \n",
       "momentum                                              701   2.8051  0.3670  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Summarize time-series means of Fama-Macbeth risk premiums\n",
    "caption, df = [\"Fama-MacBeth Cross-sectional Regression Risk Premiums\", out[0]]\n",
    "df['tvalue'] = df['mean']/df['stderr']\n",
    "df['sharpe'] = np.sqrt(12) * df['mean']/df['std']\n",
    "show(df, caption=caption)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "7732d2f3",
   "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>Benchmarks</th>\n",
       "      <th>mean</th>\n",
       "      <th>stderr</th>\n",
       "      <th>std</th>\n",
       "      <th>count</th>\n",
       "      <th>tvalue</th>\n",
       "      <th>sharpe</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Fama-French Portfolio-Sorts</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>Mom(mo)</th>\n",
       "      <td>0.0063</td>\n",
       "      <td>0.0016</td>\n",
       "      <td>0.0422</td>\n",
       "      <td>701</td>\n",
       "      <td>3.9535</td>\n",
       "      <td>0.5173</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ST_Rev(mo)</th>\n",
       "      <td>0.0047</td>\n",
       "      <td>0.0012</td>\n",
       "      <td>0.0315</td>\n",
       "      <td>701</td>\n",
       "      <td>3.9134</td>\n",
       "      <td>0.5120</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SMB(mo)</th>\n",
       "      <td>0.0020</td>\n",
       "      <td>0.0012</td>\n",
       "      <td>0.0307</td>\n",
       "      <td>701</td>\n",
       "      <td>1.7260</td>\n",
       "      <td>0.2258</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>HML(mo)</th>\n",
       "      <td>0.0030</td>\n",
       "      <td>0.0011</td>\n",
       "      <td>0.0299</td>\n",
       "      <td>701</td>\n",
       "      <td>2.6371</td>\n",
       "      <td>0.3450</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Benchmarks                     mean  stderr     std  count  tvalue  sharpe\n",
       "Fama-French Portfolio-Sorts                                               \n",
       "Mom(mo)                      0.0063  0.0016  0.0422    701  3.9535  0.5173\n",
       "ST_Rev(mo)                   0.0047  0.0012  0.0315    701  3.9134  0.5120\n",
       "SMB(mo)                      0.0020  0.0012  0.0307    701  1.7260  0.2258\n",
       "HML(mo)                      0.0030  0.0011  0.0299    701  2.6371  0.3450"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Summarize time-series means of Fama-French portfolio-sort returns\n",
    "caption, df = [\"Fama-French Portfolio-Sorts\", out[2]]\n",
    "df['tvalue'] = df['mean']/df['stderr']\n",
    "df['sharpe'] = np.sqrt(12) * df['mean']/df['std']\n",
    "show(df, caption=caption)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "863df347",
   "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>reversal</th>\n",
       "      <th>value</th>\n",
       "      <th>size</th>\n",
       "      <th>momentum</th>\n",
       "      <th>Mom(mo)</th>\n",
       "      <th>ST_Rev(mo)</th>\n",
       "      <th>SMB(mo)</th>\n",
       "      <th>HML(mo)</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Correlation of FM Risk Premiums and FF Portfolio-Sort Returns</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>reversal</th>\n",
       "      <td>1.0000</td>\n",
       "      <td>-0.0039</td>\n",
       "      <td>0.0881</td>\n",
       "      <td>-0.4460</td>\n",
       "      <td>-0.3965</td>\n",
       "      <td>0.7942</td>\n",
       "      <td>0.1340</td>\n",
       "      <td>0.0551</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>value</th>\n",
       "      <td>-0.0039</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>-0.1993</td>\n",
       "      <td>-0.1860</td>\n",
       "      <td>-0.1618</td>\n",
       "      <td>-0.0399</td>\n",
       "      <td>-0.2487</td>\n",
       "      <td>0.8178</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>size</th>\n",
       "      <td>0.0881</td>\n",
       "      <td>-0.1993</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>-0.0179</td>\n",
       "      <td>0.1198</td>\n",
       "      <td>0.0176</td>\n",
       "      <td>0.5747</td>\n",
       "      <td>-0.1421</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>momentum</th>\n",
       "      <td>-0.4460</td>\n",
       "      <td>-0.1860</td>\n",
       "      <td>-0.0179</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>0.8860</td>\n",
       "      <td>-0.2729</td>\n",
       "      <td>-0.0416</td>\n",
       "      <td>-0.2076</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Mom(mo)</th>\n",
       "      <td>-0.3965</td>\n",
       "      <td>-0.1618</td>\n",
       "      <td>0.1198</td>\n",
       "      <td>0.8860</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>-0.3057</td>\n",
       "      <td>-0.0282</td>\n",
       "      <td>-0.2053</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ST_Rev(mo)</th>\n",
       "      <td>0.7942</td>\n",
       "      <td>-0.0399</td>\n",
       "      <td>0.0176</td>\n",
       "      <td>-0.2729</td>\n",
       "      <td>-0.3057</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>0.1754</td>\n",
       "      <td>0.0206</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SMB(mo)</th>\n",
       "      <td>0.1340</td>\n",
       "      <td>-0.2487</td>\n",
       "      <td>0.5747</td>\n",
       "      <td>-0.0416</td>\n",
       "      <td>-0.0282</td>\n",
       "      <td>0.1754</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>-0.1750</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>HML(mo)</th>\n",
       "      <td>0.0551</td>\n",
       "      <td>0.8178</td>\n",
       "      <td>-0.1421</td>\n",
       "      <td>-0.2076</td>\n",
       "      <td>-0.2053</td>\n",
       "      <td>0.0206</td>\n",
       "      <td>-0.1750</td>\n",
       "      <td>1.0000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                    reversal   value    size   \n",
       "Correlation of FM Risk Premiums and FF Portfoli...                             \n",
       "reversal                                              1.0000 -0.0039  0.0881  \\\n",
       "value                                                -0.0039  1.0000 -0.1993   \n",
       "size                                                  0.0881 -0.1993  1.0000   \n",
       "momentum                                             -0.4460 -0.1860 -0.0179   \n",
       "Mom(mo)                                              -0.3965 -0.1618  0.1198   \n",
       "ST_Rev(mo)                                            0.7942 -0.0399  0.0176   \n",
       "SMB(mo)                                               0.1340 -0.2487  0.5747   \n",
       "HML(mo)                                               0.0551  0.8178 -0.1421   \n",
       "\n",
       "                                                    momentum  Mom(mo)   \n",
       "Correlation of FM Risk Premiums and FF Portfoli...                      \n",
       "reversal                                             -0.4460  -0.3965  \\\n",
       "value                                                -0.1860  -0.1618   \n",
       "size                                                 -0.0179   0.1198   \n",
       "momentum                                              1.0000   0.8860   \n",
       "Mom(mo)                                               0.8860   1.0000   \n",
       "ST_Rev(mo)                                           -0.2729  -0.3057   \n",
       "SMB(mo)                                              -0.0416  -0.0282   \n",
       "HML(mo)                                              -0.2076  -0.2053   \n",
       "\n",
       "                                                    ST_Rev(mo)  SMB(mo)   \n",
       "Correlation of FM Risk Premiums and FF Portfoli...                        \n",
       "reversal                                                0.7942   0.1340  \\\n",
       "value                                                  -0.0399  -0.2487   \n",
       "size                                                    0.0176   0.5747   \n",
       "momentum                                               -0.2729  -0.0416   \n",
       "Mom(mo)                                                -0.3057  -0.0282   \n",
       "ST_Rev(mo)                                              1.0000   0.1754   \n",
       "SMB(mo)                                                 0.1754   1.0000   \n",
       "HML(mo)                                                 0.0206  -0.1750   \n",
       "\n",
       "                                                    HML(mo)  \n",
       "Correlation of FM Risk Premiums and FF Portfoli...           \n",
       "reversal                                             0.0551  \n",
       "value                                                0.8178  \n",
       "size                                                -0.1421  \n",
       "momentum                                            -0.2076  \n",
       "Mom(mo)                                             -0.2053  \n",
       "ST_Rev(mo)                                           0.0206  \n",
       "SMB(mo)                                             -0.1750  \n",
       "HML(mo)                                              1.0000  "
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Show correlation of returns\n",
    "df = pd.concat([out[1].join(out[4]), out[4].T.join(out[3])], axis=0)\n",
    "show(df, \n",
    "     caption='Correlation of FM Risk Premiums and FF Portfolio-Sort Returns')"
   ]
  }
 ],
 "metadata": {
  "jupytext": {
   "cell_metadata_filter": "-all",
   "main_language": "python",
   "notebook_metadata_filter": "-all"
  },
  "kernelspec": {
   "display_name": "env3.11",
   "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": "3.11.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}