Standardised Bloomberg Data Processing.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Project Objectives:\n",
    "1. Use standard Bloomberg OIS and IRS data to bootstrap swap curves\n",
    "2. Calibrate swaption using SABR\n",
    "3. CMS convexity correction\n",
    "5. Extend the functions above to any standardised Bloomberg data\n",
    "6. Use relavant codes to export data back to excel"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import math\n",
    "import datetime\n",
    "import warnings\n",
    "from scipy import optimize"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# Load OIS data with daycount act/360\n",
    "# The data in use are all standardised from BGN Terminal\n",
    "# BGN shows 10-year OIS. For data beyond 10 years we use IRS quotes \n",
    "d_ois = pd.read_excel('data 3.11.xlsx',sheet_name = 'OIS')[0:18]\n",
    "d_irs = pd.read_excel('data 3.11.xlsx',sheet_name = 'IRS')[0:23] \n",
    "frame = [d_ois,d_irs[15:]]\n",
    "d_ois = pd.concat(frame)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "datetime.datetime(2019, 3, 11, 0, 0)"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Define datetime object today according to data for later use\n",
    "today = datetime.datetime.strptime(d_ois['Update'].iloc[-1],'%m/%d/%y')\n",
    "last_day = datetime.datetime(today.year,12,31)\n",
    "today "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "scrolled": false
   },
   "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>Tenor</th>\n",
       "      <th>CUSIP</th>\n",
       "      <th>Yield</th>\n",
       "      <th>Source</th>\n",
       "      <th>Tenor Days</th>\n",
       "      <th>Maturity Date</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1D</td>\n",
       "      <td>FEDL01        Index</td>\n",
       "      <td>2.40000</td>\n",
       "      <td></td>\n",
       "      <td>1</td>\n",
       "      <td>2019-03-12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2W</td>\n",
       "      <td>USSO2Z   CMPN Curncy</td>\n",
       "      <td>2.40055</td>\n",
       "      <td>CMPN</td>\n",
       "      <td>14</td>\n",
       "      <td>2019-03-25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1M</td>\n",
       "      <td>USSOA    CMPN Curncy</td>\n",
       "      <td>2.40000</td>\n",
       "      <td>CMPN</td>\n",
       "      <td>31</td>\n",
       "      <td>2019-04-11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>3M</td>\n",
       "      <td>USSOC    CMPN Curncy</td>\n",
       "      <td>2.40735</td>\n",
       "      <td>CMPN</td>\n",
       "      <td>92</td>\n",
       "      <td>2019-06-11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>5M</td>\n",
       "      <td>USSOE    CMPN Curncy</td>\n",
       "      <td>2.40565</td>\n",
       "      <td>CMPN</td>\n",
       "      <td>153</td>\n",
       "      <td>2019-08-11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>9M</td>\n",
       "      <td>USSOI    CMPN Curncy</td>\n",
       "      <td>2.41065</td>\n",
       "      <td>CMPN</td>\n",
       "      <td>275</td>\n",
       "      <td>2019-12-11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>18M</td>\n",
       "      <td>USSO1F   CMPN Curncy</td>\n",
       "      <td>2.37105</td>\n",
       "      <td>CMPN</td>\n",
       "      <td>550</td>\n",
       "      <td>2020-09-11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>3Y</td>\n",
       "      <td>USSO3    CMPN Curncy</td>\n",
       "      <td>2.27400</td>\n",
       "      <td>CMPN</td>\n",
       "      <td>1096</td>\n",
       "      <td>2022-03-11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>5Y</td>\n",
       "      <td>USSO5    CMPN Curncy</td>\n",
       "      <td>2.24600</td>\n",
       "      <td>CMPN</td>\n",
       "      <td>1827</td>\n",
       "      <td>2024-03-11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>11Y</td>\n",
       "      <td>USSWAP11 BGN Curncy</td>\n",
       "      <td>2.69078</td>\n",
       "      <td>BGN</td>\n",
       "      <td>4018</td>\n",
       "      <td>2030-03-11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>15Y</td>\n",
       "      <td>USSWAP15 BGN Curncy</td>\n",
       "      <td>2.77430</td>\n",
       "      <td>BGN</td>\n",
       "      <td>5479</td>\n",
       "      <td>2034-03-11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>25Y</td>\n",
       "      <td>USSWAP25 BGN Curncy</td>\n",
       "      <td>2.82890</td>\n",
       "      <td>BGN</td>\n",
       "      <td>9132</td>\n",
       "      <td>2044-03-11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>40Y</td>\n",
       "      <td>USSWAP40 BGN Curncy</td>\n",
       "      <td>2.80860</td>\n",
       "      <td>BGN</td>\n",
       "      <td>14610</td>\n",
       "      <td>2059-03-11</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Tenor                 CUSIP    Yield Source  Tenor Days Maturity Date\n",
       "0     1D   FEDL01        Index  2.40000                  1    2019-03-12\n",
       "2     2W  USSO2Z   CMPN Curncy  2.40055   CMPN          14    2019-03-25\n",
       "4     1M  USSOA    CMPN Curncy  2.40000   CMPN          31    2019-04-11\n",
       "6     3M  USSOC    CMPN Curncy  2.40735   CMPN          92    2019-06-11\n",
       "8     5M  USSOE    CMPN Curncy  2.40565   CMPN         153    2019-08-11\n",
       "10    9M  USSOI    CMPN Curncy  2.41065   CMPN         275    2019-12-11\n",
       "12   18M  USSO1F   CMPN Curncy  2.37105   CMPN         550    2020-09-11\n",
       "14    3Y  USSO3    CMPN Curncy  2.27400   CMPN        1096    2022-03-11\n",
       "16    5Y  USSO5    CMPN Curncy  2.24600   CMPN        1827    2024-03-11\n",
       "15   11Y   USSWAP11 BGN Curncy  2.69078    BGN        4018    2030-03-11\n",
       "17   15Y   USSWAP15 BGN Curncy  2.77430    BGN        5479    2034-03-11\n",
       "19   25Y   USSWAP25 BGN Curncy  2.82890    BGN        9132    2044-03-11\n",
       "21   40Y   USSWAP40 BGN Curncy  2.80860    BGN       14610    2059-03-11"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Date time functions\n",
    "# Get the next business day\n",
    "def payment_date(date, # The maturity date of the contract, a datetime object\n",
    "                ):\n",
    "    today = date.isoweekday()\n",
    "    # The contract matures on a business day and no change is needed\n",
    "    if today<=5:\n",
    "        return date\n",
    "    # The contract matures on a non-working day (doesn't consider public holiday)\n",
    "    else: \n",
    "        p_d = date+datetime.timedelta(days=7-today+1)\n",
    "        return p_d\n",
    "\n",
    "#Get tenor in days for OIS with act/360 daycount, this function returns int\n",
    "#This function is coded to solve standised Bloomberg quote (1d, 1w, 2w, 1m...)\n",
    "#This function can be applied to both OIS and IRS\n",
    "def get_tenor(string):\n",
    "    ret = 0\n",
    "    if 'D' in string: #1D\n",
    "        maturity = today + datetime.timedelta(days=1)\n",
    "        ret = 1\n",
    "        \n",
    "    elif 'W' in string: #1W, 2W, 3W\n",
    "        number = int(string.replace('W',''))*7\n",
    "        maturity = today + datetime.timedelta(days=number)\n",
    "        ret = number\n",
    "\n",
    "    elif 'M' in string: #1M, 2M, 3M, 4M, 5M, 6M, 9M, 18M\n",
    "        number = int(string.replace('M',''))\n",
    "        mtye = last_day.month-today.month # Months till year end\n",
    "        if mtye < number: # The contract doesn't end in same year\n",
    "            if mtye+12 < number: # The contract spans across 3 years\n",
    "                maturity = datetime.datetime(today.year+2,number-(mtye+12),today.day)\n",
    "            else: # The contract ends next year\n",
    "                maturity = datetime.datetime(today.year+1,number-mtye,today.day)\n",
    "        else: # The contract ends in the same year\n",
    "            maturity = datetime.datetime(today.year,today.month+number,today.day)\n",
    "        ret = (maturity - today).days\n",
    "        \n",
    "    else: # 1, 2, 3, 4, 5, 10Y\n",
    "        y_d = int(string.replace('Y',''))\n",
    "        maturity = datetime.datetime(today.year+y_d,today.month,today.day)\n",
    "        ret = (maturity - today).days\n",
    "    return ret\n",
    "\n",
    "# Polish the Dataframe\n",
    "d_ois['Tenor Days'] = d_ois['Tenor'].apply(get_tenor)\n",
    "d_ois['Tenor Years'] = d_ois['Tenor Days']/360 # act/360\n",
    "d_ois['Maturity Date'] = d_ois['Tenor Days'].apply(datetime.timedelta)+today\n",
    "d_ois['Payment Date'] = d_ois['Maturity Date'].apply(payment_date)\n",
    "d_ois.iloc[::2,[0,1,3,4,6,8]]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Matching the OIS Quotes:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "By PV Floating = PV Fixed :\n",
    "$$\n",
    "\\Pi_{i=1}^{N}(1+\\Delta_{i-1}·\\frac{1}{\\Delta_{i-1}}·\\frac{D_o(0,t_{i-1})-D_o(0,t_i)}{D_o(0,t_i)})-1 = OIS_t \\Delta_t \\sum_{i=1}^{t}D_0(t_i)$$\n",
    "\n",
    "$$\n",
    "\\Pi_{i=1}^{N}\\frac{D_o(0,t_{i-1})}{D_o(0,t_i)}-1 = OIS_t \\Delta_t \\sum_{i=1}^{t}D_0(t_i)\n",
    "$$\n",
    "Hence we get for single payment contracts (ends within one year):\n",
    "$$\n",
    "\\frac{1}{D_o(0,t_N)}-1 = OIS_t \\Delta_t \\sum_{i=1}^{t}D_0(t_i)\n",
    "$$\n",
    "For multiple payment contracts:\n",
    "\n",
    "$$\n",
    "D(0,T_i)\\Pi_{i=1}^{365}(1+\\delta f) = D(0,T_{i-1})\n",
    "$$\n",
    "\n",
    "$$\n",
    "D(0,T_i)\\Pi_{i=1}^{365}(1+\\delta f)-D(0,T_i)= D(0,T_{i-1}) - D(0,T_i)\n",
    "$$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# All global data are initially defined using this cell\n",
    "# 1. Actual days in year, 2. Get PVBP \n",
    "# 3. Bisection search of dfs that matches all observable contract tenors \n",
    "# 4. Calculate dfs of each year\n",
    "# BGN shows OIS quotes for ten years and from 11th years onwards IRS is used\n",
    "\n",
    "# Here we use a separate list to store all the dfs for OIS quotes table\n",
    "odfs_t = np.zeros(17)\n",
    "\n",
    "#This is the list of 3m time interval odfs \n",
    "len_dfs = int(d_irs['Tenor'].iloc[-1].replace('Y',''))*4 #len = 200\n",
    "odfs = np.zeros(len_dfs)\n",
    "\n",
    "#odfs_y is a list of yearly discount factors\n",
    "odfs_y = np.append(odfs[3::4],[0])\n",
    "\n",
    "# 1.How many actual days in one year\n",
    "def d_in_y(T# Tenor\n",
    "          ):\n",
    "    doy = np.zeros(T+1) # days of years, the days of first year is at index 1\n",
    "    for i in range(1,T+1):\n",
    "        maturity = datetime.datetime(today.year+i,today.month,today.day)\n",
    "        doy[i] = int((maturity-today).days)-sum(doy[:i])\n",
    "    return doy\n",
    "doy = d_in_y(len_dfs+1)\n",
    "\n",
    "#2. Calculate the PVBP for an multi-paynement OIS contract\n",
    "def PVBP(T, # Contract tenor\n",
    "         guess, # initial guess\n",
    "         daycount ='ACT/360'):\n",
    "    ret = np.zeros(T+1) # of length T+1\n",
    "    diy = doy[:T+1] # of length T+1\n",
    "    copy = odfs_y.copy()\n",
    "    # Renew the dfs with intepolation\n",
    "    fu = 1 #first unknown df in odfs_y\n",
    "    while(odfs_y[fu]!=0):\n",
    "        fu+=1\n",
    "    if fu==T: # No interpolation needed\n",
    "        ret = np.append(copy[:T],guess) #retrun is all the known dfs and the guess\n",
    "    else:\n",
    "        # Known dfs,\n",
    "        k_d = copy[:fu-1]\n",
    "        # Unknown part of PVBP, solved with interpolation\n",
    "        u_d = pd.Series(np.linspace(copy[fu-1],guess,T-fu+1+1))\n",
    "        ret = np.append(k_d,u_d)\n",
    "    if daycount == 'ACT/360':\n",
    "        ret = ret*pd.Series(diy/360) #ret is of length T+1\n",
    "    else:\n",
    "        ret = pd.Series(odfs_y[:T+1])\n",
    "    return ret\n",
    "\n",
    "# For Tenor less than 1 year, only one payment is needed\n",
    "# For 18M contract, the first payment is in 6-month and the second is on maturity\n",
    "\n",
    "#3. This function uses bisection to find the OIS for every contract except 18M one\n",
    "def bi_oisdf(s, #swap rate\n",
    "           T, #Tenor of contract. T is a string obatined from the Tenor column\n",
    "           m='s'#s is for single payment, m is for multiple payment\n",
    "            ):\n",
    "    # single payment, solve for OIS*dt*D^2 = 1 - D\n",
    "    tol = 1e-10\n",
    "    ret = None\n",
    "    # ACT/360 single payment\n",
    "    if m=='s':\n",
    "        ll, ul= 0.8, 1 # lower limit, upper limit \n",
    "        dt = get_tenor(T)/360\n",
    "        diff = lambda x:s*dt*x**2+x-1\n",
    "        ret = optimize.bisect(diff,ll,ul,xtol=tol)\n",
    "    # multiple payments \n",
    "    else:\n",
    "        ll, ul, diff = 0.1, 1, 0 # lower limit, upper limit \n",
    "        ten = int(T.replace('Y',''))\n",
    "        diff = lambda x:s*sum(PVBP(ten,x))+x-1\n",
    "        ret = optimize.bisect(diff,ll,ul,xtol=tol)\n",
    "    return ret\n",
    "\n",
    "#4. Get the Discount Factors from the multi-payment OIS quotes by applying bi_oisdf\n",
    "def OISDF(s, # swap rate\n",
    "          T, # tenor, this a string input, quoted directly from the data sheet\n",
    "          odfs_y):\n",
    "    ret = None\n",
    "    tenor_d = get_tenor(T) #tenor in days\n",
    "    # A special case where one payment is 6 month, the other is on maturity\n",
    "    if T=='18M': \n",
    "        maturity = today + datetime.timedelta(days=tenor_d)\n",
    "        s_pd = maturity # Second, first payment date\n",
    "        f_pd = datetime.datetime(s_pd.year-1,s_pd.month,s_pd.day) \n",
    "        dtf, dts = (f_pd-today).days/360, (s_pd-today).days/360\n",
    "        ret = (1-s*odf_s[9]*dtf)/(1+s*dts)\n",
    "    # Annual payments for multiple years\n",
    "    else:\n",
    "        i_s = int(T.replace('Y',''))#renew the discount factor list\n",
    "        ret = bi_oisdf(s,T,m='m')  \n",
    "    return ret"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Get the odfs for all single payment contract, including 1Y contract\n",
    "odf_s = np.zeros(12) # There are 12 single payment contracts from a single quote\n",
    "for i in range(len(odf_s)):\n",
    "    odf_s[i] = bi_oisdf(d_ois['Yield'].iloc[i]/100,d_ois['Tenor'].iloc[i])\n",
    "odfs_y[1] = odf_s[-1]\n",
    "\n",
    "# Get OIS for 18 month contract\n",
    "d_18 = OISDF(d_ois['Yield'].iloc[12]/100,d_ois['Tenor'].iloc[12],odfs_y)\n",
    "\n",
    "# OIS for multiple payment contract\n",
    "for i in range(13,26):\n",
    "    iy = int(d_ois['Tenor'].iloc[i].replace('Y',''))\n",
    "    odfs_y[iy] = OISDF(d_ois['Yield'].iloc[i]/100,d_ois['Tenor'].iloc[i],odfs_y)\n",
    "    #intepolation\n",
    "    # find the index of the first unknown\n",
    "    if i!=13:\n",
    "        fu = int(d_ois['Tenor'].iloc[i-1].replace('Y',''))+1\n",
    "    else:\n",
    "        fu = 1+1\n",
    "    inte = np.linspace(odfs_y[fu-1],odfs_y[iy],iy-fu+1+1)\n",
    "    for i1 in range(fu-1,iy+1):\n",
    "        odfs_y[i1] = inte[i1-fu+1]\n",
    "\n",
    "#plotting\n",
    "fig = plt.figure(figsize=(10,6))\n",
    "ax = fig.add_subplot(111)\n",
    "ax.plot(odfs_y[1:], 'go')\n",
    "ax.spines['top'].set_color('none')\n",
    "ax.spines['right'].set_color('none')\n",
    "ax.grid(linestyle = '--', alpha = 1)\n",
    "ax.set_title('OIS Discount Factors',fontsize=16)\n",
    "ax.set_xlabel('Tenor',fontsize=16)\n",
    "ax.set_ylabel('Discount Factor',fontsize=16)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Calculate the IRS Discount Factors"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# load data\n",
    "d_irs = pd.read_excel('data 3.11.xlsx',sheet_name = 'IRS')[0:23] "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "scrolled": false
   },
   "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>Tenor</th>\n",
       "      <th>CUSIP</th>\n",
       "      <th>Yield</th>\n",
       "      <th>Source</th>\n",
       "      <th>Tenor Days</th>\n",
       "      <th>Maturity Date</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3M</td>\n",
       "      <td>US0003M CMPN Index</td>\n",
       "      <td>2.59663</td>\n",
       "      <td>CMPN</td>\n",
       "      <td>92</td>\n",
       "      <td>2019-06-11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>9M</td>\n",
       "      <td>EDU9 BGN Comdty</td>\n",
       "      <td>2.60868</td>\n",
       "      <td>BGN</td>\n",
       "      <td>275</td>\n",
       "      <td>2019-12-11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>15M</td>\n",
       "      <td>EDH0 BGN Comdty</td>\n",
       "      <td>2.55618</td>\n",
       "      <td>BGN</td>\n",
       "      <td>458</td>\n",
       "      <td>2020-06-11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>2Y</td>\n",
       "      <td>USSWAP2 BGN Curncy</td>\n",
       "      <td>2.60220</td>\n",
       "      <td>BGN</td>\n",
       "      <td>731</td>\n",
       "      <td>2021-03-11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>4Y</td>\n",
       "      <td>USSWAP4 BGN Curncy</td>\n",
       "      <td>2.51730</td>\n",
       "      <td>BGN</td>\n",
       "      <td>1461</td>\n",
       "      <td>2023-03-11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>6Y</td>\n",
       "      <td>USSW6 BGN Curncy</td>\n",
       "      <td>2.53900</td>\n",
       "      <td>BGN</td>\n",
       "      <td>2192</td>\n",
       "      <td>2025-03-11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>8Y</td>\n",
       "      <td>USSW8 BGN Curncy</td>\n",
       "      <td>2.59600</td>\n",
       "      <td>BGN</td>\n",
       "      <td>2922</td>\n",
       "      <td>2027-03-11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>10Y</td>\n",
       "      <td>USSWAP10 BGN Curncy</td>\n",
       "      <td>2.66090</td>\n",
       "      <td>BGN</td>\n",
       "      <td>3653</td>\n",
       "      <td>2029-03-11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>12Y</td>\n",
       "      <td>USSWAP12 BGN Curncy</td>\n",
       "      <td>2.71830</td>\n",
       "      <td>BGN</td>\n",
       "      <td>4383</td>\n",
       "      <td>2031-03-11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>20Y</td>\n",
       "      <td>USSWAP20 BGN Curncy</td>\n",
       "      <td>2.81790</td>\n",
       "      <td>BGN</td>\n",
       "      <td>7305</td>\n",
       "      <td>2039-03-11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>30Y</td>\n",
       "      <td>USSWAP30 BGN Curncy</td>\n",
       "      <td>2.83110</td>\n",
       "      <td>BGN</td>\n",
       "      <td>10958</td>\n",
       "      <td>2049-03-11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>50Y</td>\n",
       "      <td>USSWAP50 BGN Curncy</td>\n",
       "      <td>2.77175</td>\n",
       "      <td>BGN</td>\n",
       "      <td>18263</td>\n",
       "      <td>2069-03-11</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Tenor                CUSIP    Yield Source  Tenor Days Maturity Date\n",
       "0     3M   US0003M CMPN Index  2.59663   CMPN          92    2019-06-11\n",
       "2     9M      EDU9 BGN Comdty  2.60868    BGN         275    2019-12-11\n",
       "4    15M      EDH0 BGN Comdty  2.55618    BGN         458    2020-06-11\n",
       "6     2Y   USSWAP2 BGN Curncy  2.60220    BGN         731    2021-03-11\n",
       "8     4Y   USSWAP4 BGN Curncy  2.51730    BGN        1461    2023-03-11\n",
       "10    6Y     USSW6 BGN Curncy  2.53900    BGN        2192    2025-03-11\n",
       "12    8Y     USSW8 BGN Curncy  2.59600    BGN        2922    2027-03-11\n",
       "14   10Y  USSWAP10 BGN Curncy  2.66090    BGN        3653    2029-03-11\n",
       "16   12Y  USSWAP12 BGN Curncy  2.71830    BGN        4383    2031-03-11\n",
       "18   20Y  USSWAP20 BGN Curncy  2.81790    BGN        7305    2039-03-11\n",
       "20   30Y  USSWAP30 BGN Curncy  2.83110    BGN       10958    2049-03-11\n",
       "22   50Y  USSWAP50 BGN Curncy  2.77175    BGN       18263    2069-03-11"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def get_tenor_irs(string):\n",
    "    ret = None\n",
    "    if 'M' in string:\n",
    "        ret = int(string.replace('M',''))/12\n",
    "    else:\n",
    "        ret = int(string.replace('Y',''))\n",
    "    return ret\n",
    "\n",
    "#Get time delta in years for 30/360 and act/360, this function returns floats\n",
    "d_irs['Tenor Days'] = d_irs['Tenor'].apply(get_tenor)\n",
    "d_irs['Tenor Years'] = d_irs['Tenor Days']/360 # act/360\n",
    "d_irs['Maturity Date'] = d_irs['Tenor Days'].apply(datetime.timedelta)+today\n",
    "d_irs['Payment Date'] = d_irs['Maturity Date'].apply(payment_date)\n",
    "d_irs['Fixed Leg Dt'] = d_irs['Tenor'].apply(get_tenor_irs)\n",
    "d_irs.iloc[::2,[0,1,3,4,6,8]]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Although the two legs use differnent daycount convention, their payments happen in the same day and both legs use OIS discounting\n",
    "\n",
    "The Floating leg is of quarterly Libor payment with daycount Act/360:\n",
    "$$\n",
    "    D(0,T_i) \\Delta_{3m,act/360}L(T_{i-1},T_i)= \\sum_{i=1}^{n}D(0,T_{i-1})-D(0,T_{i})\n",
    "$$\n",
    "where:\n",
    "$$\n",
    "\\sum_{i=1}^{n}D(0,T_{i-1})-D(0,T_{i}) = 1- D(0,T_i)\n",
    "$$\n",
    "\n",
    "The Fixed leg is of Semi-Annual payment with daycount 30/360, following Treasury Bill:\n",
    "$$\n",
    "IRS_t \\Delta_{6m,30/360} \\sum_{i=1}^{n}D(0,T_i)\n",
    "$$\n",
    "\n",
    "By PV Fixed = PV Floating:\n",
    "$$\n",
    "IRS_t \\Delta_{6m,30/360} \\sum_{i=1}^{n}D(0,T_i) = 1- D(0,T_i)\n",
    "$$\n",
    "Here $\\Delta_t\\sum_{i=1}^{t}D(0,t_i)$ is the denominator of the par IRS swap rate, and can be interpreted as PVBP (Present Value of a Basis Point): $$P_{1,n} = \\Delta\\sum_{i=1}^{t}D(0,T_i)$$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# For IRS, the market practice is also OIS discounting\n",
    "# 1. odfs is the list of 3m OIS discount factors, length 200\n",
    "odfs3m = np.zeros(len_dfs)\n",
    "odfs6m = np.zeros(int(len_dfs/2))\n",
    "\n",
    "# 2. Get the actual days for every 3 month period\n",
    "# This will be used when calculating implied forward Libor\n",
    "di3m = np.zeros(len_dfs)\n",
    "last_pd = today\n",
    "next_pd = today\n",
    "for i in range(len_dfs):\n",
    "    if last_pd.month <= 9: # The contract ends in the same year\n",
    "        next_pd = datetime.datetime(last_pd.year,last_pd.month+3,last_pd.day)\n",
    "    else: # The contract ends next year\n",
    "        next_pd = datetime.datetime(last_pd.year+1,last_pd.month+3-12,last_pd.day)\n",
    "    di3m[i] =  (next_pd-last_pd).days\n",
    "    last_pd = next_pd\n",
    "\n",
    "# 3. Calculation\n",
    "# 3M to 18M contracts have 1 payment, no interpolation needed\n",
    "for i in range(6):\n",
    "    if i==0:\n",
    "        k_d = 0 # known dfs\n",
    "    else:\n",
    "        k_d = sum(odfs3m[:i])\n",
    "    irs = d_irs['Yield'].iloc[i]/100\n",
    "    dt = 90/360\n",
    "    odfs3m[i] = (1-irs*dt*k_d)/(1+irs*dt)\n",
    "\n",
    "\n",
    "odfs6m[0] ,odfs6m[1], odfs6m[2] = odfs3m[1], odfs3m[3], odfs3m[5]\n",
    "\n",
    "#There are 23 observable contracts on the market\n",
    "# From 18M onwards, only yearly contracts are quoted and interpolation is needed\n",
    "len_irs = d_irs.index.values[-1]+1\n",
    "\n",
    "# Get the 2 year contract first\n",
    "s = d_irs['Yield'].iloc[6]/100\n",
    "odfs6m[3] = (1-0.5*s*sum(odfs6m))/(1+0.5*s)\n",
    "odfs3m[7] = odfs6m[3]\n",
    "odfs3m[6] = (odfs3m[7]+odfs3m[5])/2\n",
    "\n",
    "# the rest\n",
    "for i in range(7,len_irs):\n",
    "    s = d_irs['Yield'].iloc[i]/100 # par swap\n",
    "    dt = 180/360  # Semi-annual payment\n",
    "    \n",
    "    # find the first unknown df\n",
    "    fu = 4\n",
    "    while(odfs6m[fu]!=0):\n",
    "        fu+=1\n",
    "    k_d = odfs6m[:fu] #sum of known dfs, of length fu-1\n",
    "    # Last unknown, the index of the last df of the contract\n",
    "    lu = get_tenor_irs(d_irs['Tenor'].iloc[i])*2-1\n",
    "    # Use bisection to get the df\n",
    "    tol = 1e-10\n",
    "    ll, ul= 0.1, 1 # lower limit, upper limit\n",
    "    def diff(x):\n",
    "        u_d = np.linspace(odfs6m[fu-1],x,lu-fu+1+1)\n",
    "        d = s*dt*(sum(k_d)+sum(u_d))+x-1\n",
    "        return d \n",
    "    df_solution = optimize.bisect(diff,ll,ul,xtol=tol)\n",
    "    u_d = np.linspace(odfs6m[fu-1],df_solution,lu-fu+1+1)\n",
    "    for i1 in range(fu,lu+1):\n",
    "        odfs6m[i1] = u_d[i1-fu+1]\n",
    "    # Filling the odfs3m by interpolation\n",
    "    df3m = np.linspace(u_d[0],u_d[-1],(lu-fu+1)*2+1)\n",
    "    for i in range(len(df3m)):\n",
    "        odfs3m[int((fu-1)*2+1+i)] = df3m[i]\n",
    "        \n",
    "#plotting\n",
    "fig = plt.figure(figsize=(10,6))\n",
    "ax = fig.add_subplot(111)\n",
    "ax.plot(odfs3m[3::4], 'go')\n",
    "ax.spines['top'].set_color('none')\n",
    "ax.spines['right'].set_color('none')\n",
    "ax.grid(linestyle = '--', alpha = 1)\n",
    "ax.set_title('IRS Discount Factors',fontsize=16)\n",
    "ax.set_xlabel('Tenor',fontsize=16)\n",
    "ax.set_ylabel('Discount Factor',fontsize=16)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Get 3-month Implied Libor\n",
    "imp_lib = np.zeros(len_dfs)\n",
    "for i in range(len_dfs):\n",
    "    if i!=0:\n",
    "        imp_lib[i] = (odfs3m[i-1]-odfs3m[i])/odfs3m[i]/(di3m[i]/360)\n",
    "    else:\n",
    "        imp_lib[0] = (1-odfs3m[0])/odfs3m[0]/(di3m[0]/360)\n",
    "        \n",
    "#plotting\n",
    "fig = plt.figure(figsize=(10,6))\n",
    "ax = fig.add_subplot(111)\n",
    "ax.plot(imp_lib[3::4], 'bo', lw=20)\n",
    "ax.spines['top'].set_color('none')\n",
    "ax.spines['right'].set_color('none')\n",
    "ax.grid(linestyle = '--', alpha = 1)\n",
    "ax.set_title('Implied 3M Forward Libor',fontsize=16)\n",
    "ax.set_xlabel('Payment Date',fontsize=16)\n",
    "ax.set_ylabel('Forward Libor',fontsize=16)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Calculate Forward Swap Rates:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The Floating Leg is of quarterly payment by ACT/360 daycount:\n",
    "$$\n",
    "\\sum_{i=e}^{e+t}\\Delta_{3m,ACT/360}L(T_{i},T_{i+1})D(0,T_{i+1})\n",
    "$$\n",
    "The Fixed Leg is of semi-annual payment by 30/360 daycount, or PVBP:\n",
    "$$\n",
    "\\sum_{i=e}^{e+t}\\Delta_{6m,30/360}D(0,T_{i+1})\n",
    "$$\n",
    "Hence, the Forward Swap Rate can be calculated by Floating Leg divided by the Fixed Leg:\n",
    "$$\n",
    "IRS_{Forward} = \\frac{\\sum_{i=e}^{e+t}\\Delta_{3m,ACT/360}L(T_{i},T_{i+1})D(0,T_{i+1})}{\\sum_{i=e}^{e+t}\\Delta_{6m,30/360}D(0,T_{i+1})}\n",
    "$$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "scrolled": false
   },
   "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>3Mo x</th>\n",
       "      <th>1Yr x</th>\n",
       "      <th>2Yr x</th>\n",
       "      <th>5Yr x</th>\n",
       "      <th>10Yr x</th>\n",
       "      <th>20Yr x</th>\n",
       "      <th>30Yr x</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2Yr</th>\n",
       "      <td>0.027326</td>\n",
       "      <td>0.027623</td>\n",
       "      <td>0.027899</td>\n",
       "      <td>0.028808</td>\n",
       "      <td>0.028660</td>\n",
       "      <td>0.031362</td>\n",
       "      <td>0.027632</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5Yr</th>\n",
       "      <td>0.028179</td>\n",
       "      <td>0.028875</td>\n",
       "      <td>0.028666</td>\n",
       "      <td>0.030590</td>\n",
       "      <td>0.027961</td>\n",
       "      <td>0.028676</td>\n",
       "      <td>0.029043</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10Yr</th>\n",
       "      <td>0.028724</td>\n",
       "      <td>0.026685</td>\n",
       "      <td>0.028023</td>\n",
       "      <td>0.029135</td>\n",
       "      <td>0.029131</td>\n",
       "      <td>0.028511</td>\n",
       "      <td>0.030329</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20Yr</th>\n",
       "      <td>0.030422</td>\n",
       "      <td>0.030089</td>\n",
       "      <td>0.029304</td>\n",
       "      <td>0.028642</td>\n",
       "      <td>0.027293</td>\n",
       "      <td>0.028458</td>\n",
       "      <td>0.028256</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30Yr</th>\n",
       "      <td>0.027547</td>\n",
       "      <td>0.026700</td>\n",
       "      <td>0.024059</td>\n",
       "      <td>0.024959</td>\n",
       "      <td>0.026620</td>\n",
       "      <td>0.025554</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         3Mo x     1Yr x     2Yr x     5Yr x    10Yr x    20Yr x    30Yr x\n",
       "2Yr   0.027326  0.027623  0.027899  0.028808  0.028660  0.031362  0.027632\n",
       "5Yr   0.028179  0.028875  0.028666  0.030590  0.027961  0.028676  0.029043\n",
       "10Yr  0.028724  0.026685  0.028023  0.029135  0.029131  0.028511  0.030329\n",
       "20Yr  0.030422  0.030089  0.029304  0.028642  0.027293  0.028458  0.028256\n",
       "30Yr  0.027547  0.026700  0.024059  0.024959  0.026620  0.025554       NaN"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# BGN shows quotes for upto 50 years \n",
    "#we can only calibrate for contracts as far as 30x20\n",
    "d_ir = pd.read_excel('data 3.11.xlsx',sheet_name='IR Swaption')[:-1]\n",
    "\n",
    "# Get Expiry and Tenor separate\n",
    "ir_q = d_ir['Expiry x Tenor']\n",
    "exp, ten = [], []\n",
    "len_ir = d_ir.index.values[-1]+1\n",
    "for i in range(len_ir):\n",
    "    spl = ir_q[i].split('x')\n",
    "    if 'Mo' in spl[0]:\n",
    "        exp.append(int(spl[0].replace('Mo',''))/12)\n",
    "    else:\n",
    "        exp.append(int(spl[0].replace('Yr','')))\n",
    "    ten.append(int(spl[1].replace('Yr','')))\n",
    "    \n",
    "# Calculate the forwards\n",
    "def fir(expiry,\n",
    "        tenor):\n",
    "    start_con = int(expiry*4) # contract of start\n",
    "    len_con = int((expiry+tenor)*4) # length of contract\n",
    "    sum_flt, sum_fix = 0, 0\n",
    "    # Floating Leg\n",
    "    for i in range(start_con,len_con):\n",
    "        dt_flt = di3m[i]/360\n",
    "        sum_flt += dt_flt*imp_lib[i]*odfs3m[i]\n",
    "    # Fixed Leg\n",
    "    dt_fix = 0.5 \n",
    "    odf = odfs3m[start_con+1:len_con:2].copy()\n",
    "    for i in range(len(odf)):\n",
    "        sum_fix += dt_fix*odf[i]\n",
    "    return sum_flt/sum_fix\n",
    "firs = []\n",
    "for i in range(len_ir):\n",
    "    firs.append(fir(exp[i],ten[i]))\n",
    "\n",
    "#Plotting\n",
    "firs_df = firs.copy()\n",
    "firs_df.append(math.nan)\n",
    "firs_df = np.array(firs_df).reshape(5,7)\n",
    "x_lab = ['3Mo x','1Yr x','2Yr x','5Yr x','10Yr x','20Yr x','30Yr x']\n",
    "y_lab = ['2Yr','5Yr','10Yr','20Yr','30Yr']\n",
    "pd.DataFrame(firs_df,index=y_lab,columns=x_lab)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Part II: Swaption Calibration with SABR"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "bps_range = d_ir.columns\n",
    "d_ir.columns =['Expiry x Tenor']+[-0.02, -0.01, -0.005, -0.0025] + ['ATM'] + [0.0025, 0.005, 0.01, 0.02]\n",
    "d_ir['Contract Expiry'] = exp\n",
    "d_ir['Contract Tenor'] = ten\n",
    "ext = d_ir['Expiry x Tenor'].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "def SABR(F, K, T, alpha, beta, rho, nu):\n",
    "    X = K\n",
    "    if F == K:\n",
    "        numer1 = (((1 - beta)**2)/24)*alpha*alpha/(F**(2 - 2*beta))\n",
    "        numer2 = 0.25*rho*beta*nu*alpha/(F**(1 - beta))\n",
    "        numer3 = ((2 - 3*rho*rho)/24)*nu*nu\n",
    "        VolAtm = alpha*(1 + (numer1 + numer2 + numer3)*T)/(F**(1-beta))\n",
    "        sabrsigma = VolAtm\n",
    "    else:\n",
    "        z = (nu/alpha)*((F*X)**(0.5*(1-beta)))*np.log(F/X)\n",
    "        zhi = np.log((((1 - 2*rho*z + z*z)**0.5) + z - rho)/(1 - rho))\n",
    "        numer1 = (((1 - beta)**2)/24)*((alpha*alpha)/((F*X)**(1 - beta)))\n",
    "        numer2 = 0.25*rho*beta*nu*alpha/((F*X)**((1 - beta)/2))\n",
    "        numer3 = ((2 - 3*rho*rho)/24)*nu*nu\n",
    "        numer = alpha*(1 + (numer1 + numer2 + numer3)*T)*z\n",
    "        denom1 = ((1 - beta)**2/24)*(np.log(F/X))**2\n",
    "        denom2 = (((1 - beta)**4)/1920)*((np.log(F/X))**4)\n",
    "        denom = ((F*X)**((1 - beta)/2))*(1 + denom1 + denom2)*zhi\n",
    "        sabrsigma = numer/denom\n",
    "    return sabrsigma\n",
    "def SABR_calibration(expiry, # contract expiry exp\n",
    "                     tenor, # contract tenor ten\n",
    "                    ): \n",
    "    # Get the index of the contract\n",
    "    con_ind = 0\n",
    "    for i1 in range(len_ir):\n",
    "        if (exp[i1] == float(expiry)) & (ten[i1] == float(tenor)):\n",
    "            con_ind = i1\n",
    "            break\n",
    "            \n",
    "    # Get vols for the contract\n",
    "    con_vol = d_ir.iloc[con_ind][1:10]\n",
    "    atm_vol = con_vol[4]\n",
    "    \n",
    "    # Get vol from SABR for every bps change on the contract\n",
    "    bps_change = d_ir.columns.values[1:10]\n",
    "    bps_change[4] = 0\n",
    "    bps_change = bps_change.astype(float)\n",
    "    F = firs[con_ind]# Forward Swap Rate\n",
    "    \n",
    "    # Get the list of vol for iteration\n",
    "    def get_vol_lst(guess):\n",
    "        vol_lst = []\n",
    "        for bps in bps_change:\n",
    "            vol_lst.append(SABR(F,F+bps,tenor,\n",
    "                           guess[0],guess[1],guess[2],guess[3]))\n",
    "        return vol_lst\n",
    "    \n",
    "    # Use error norm to calibrate\n",
    "    sabr_calib = lambda x: np.linalg.norm(con_vol - get_vol_lst(x))\n",
    "    \n",
    "    # Calibration\n",
    "    guess = [0.5,0.5,0.5,0.5] # alpha>=0, 0<=beta<=1, -1<rho<1, nu\n",
    "    bounds = ((0,np.inf),(0,1),(-1,1),(0,np.inf)) #α > 0 , 0 ≤ β ≤ 1 , −1 ≤ ρ ≤ 1 , ν > 0\n",
    "    res = optimize.minimize(sabr_calib,guess,method='SLSQP',\n",
    "                   options={'maxiter':1000},\n",
    "                   bounds=bounds,tol=10e-10)\n",
    "    return res.x, get_vol_lst(res.x)\n",
    "\n",
    "# List of parameters and calibrated ATM vol\n",
    "para_lst = []\n",
    "ccon_vol = []\n",
    "warnings.simplefilter('ignore',category=RuntimeWarning)\n",
    "\n",
    "for i in range(len_ir):\n",
    "    res = SABR_calibration(exp[i],ten[i])\n",
    "    para_lst.append(res[0])\n",
    "    ccon_vol.append(res[1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "scrolled": true
   },
   "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>Alpha</th>\n",
       "      <th>Beta</th>\n",
       "      <th>Rho</th>\n",
       "      <th>Nu</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3Mo x 2Yr</th>\n",
       "      <td>0.039575</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-0.393743</td>\n",
       "      <td>9.000788</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3Mo x 20Yr</th>\n",
       "      <td>0.075752</td>\n",
       "      <td>0.416629</td>\n",
       "      <td>-0.166364</td>\n",
       "      <td>4.183993</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1Yr x 5Yr</th>\n",
       "      <td>0.749018</td>\n",
       "      <td>0.404722</td>\n",
       "      <td>-0.256708</td>\n",
       "      <td>0.000010</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1Yr x 30Yr</th>\n",
       "      <td>0.058774</td>\n",
       "      <td>0.225821</td>\n",
       "      <td>-0.118025</td>\n",
       "      <td>2.833496</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2Yr x 10Yr</th>\n",
       "      <td>0.104135</td>\n",
       "      <td>0.208759</td>\n",
       "      <td>-0.129472</td>\n",
       "      <td>3.807333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5Yr x 2Yr</th>\n",
       "      <td>0.313803</td>\n",
       "      <td>0.299224</td>\n",
       "      <td>-0.224705</td>\n",
       "      <td>6.173815</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5Yr x 20Yr</th>\n",
       "      <td>0.154713</td>\n",
       "      <td>0.330905</td>\n",
       "      <td>-0.171073</td>\n",
       "      <td>2.724443</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10Yr x 5Yr</th>\n",
       "      <td>0.222832</td>\n",
       "      <td>0.294026</td>\n",
       "      <td>-0.129861</td>\n",
       "      <td>4.192459</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10Yr x 30Yr</th>\n",
       "      <td>0.115605</td>\n",
       "      <td>0.295134</td>\n",
       "      <td>-0.185133</td>\n",
       "      <td>2.345944</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20Yr x 10Yr</th>\n",
       "      <td>0.172797</td>\n",
       "      <td>0.286278</td>\n",
       "      <td>-0.192540</td>\n",
       "      <td>3.083870</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30Yr x 2Yr</th>\n",
       "      <td>2.048519</td>\n",
       "      <td>0.624297</td>\n",
       "      <td>0.994454</td>\n",
       "      <td>0.155654</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30Yr x 20Yr</th>\n",
       "      <td>0.223254</td>\n",
       "      <td>0.353965</td>\n",
       "      <td>-0.298575</td>\n",
       "      <td>2.341652</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                Alpha      Beta       Rho        Nu\n",
       "3Mo x 2Yr    0.039575  0.000000 -0.393743  9.000788\n",
       "3Mo x 20Yr   0.075752  0.416629 -0.166364  4.183993\n",
       "1Yr x 5Yr    0.749018  0.404722 -0.256708  0.000010\n",
       "1Yr x 30Yr   0.058774  0.225821 -0.118025  2.833496\n",
       "2Yr x 10Yr   0.104135  0.208759 -0.129472  3.807333\n",
       "5Yr x 2Yr    0.313803  0.299224 -0.224705  6.173815\n",
       "5Yr x 20Yr   0.154713  0.330905 -0.171073  2.724443\n",
       "10Yr x 5Yr   0.222832  0.294026 -0.129861  4.192459\n",
       "10Yr x 30Yr  0.115605  0.295134 -0.185133  2.345944\n",
       "20Yr x 10Yr  0.172797  0.286278 -0.192540  3.083870\n",
       "30Yr x 2Yr   2.048519  0.624297  0.994454  0.155654\n",
       "30Yr x 20Yr  0.223254  0.353965 -0.298575  2.341652"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Show the calibrated parameters of SABR for different contracts\n",
    "c_para = pd.DataFrame(para_lst,\n",
    "                      columns=['Alpha','Beta','Rho','Nu'],\n",
    "                      index=ext)\n",
    "c_para[::3]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "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>-200bps</th>\n",
       "      <th>-100bps</th>\n",
       "      <th>ATM</th>\n",
       "      <th>100bps</th>\n",
       "      <th>200bps</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3Mo x 2Yr</th>\n",
       "      <td>69.649544</td>\n",
       "      <td>34.689795</td>\n",
       "      <td>16.708668</td>\n",
       "      <td>17.124383</td>\n",
       "      <td>21.387780</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3Mo x 20Yr</th>\n",
       "      <td>58.027905</td>\n",
       "      <td>31.340723</td>\n",
       "      <td>16.915073</td>\n",
       "      <td>21.086362</td>\n",
       "      <td>26.966016</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1Yr x 5Yr</th>\n",
       "      <td>67.079706</td>\n",
       "      <td>35.281748</td>\n",
       "      <td>25.181680</td>\n",
       "      <td>20.055387</td>\n",
       "      <td>16.904566</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1Yr x 30Yr</th>\n",
       "      <td>46.351442</td>\n",
       "      <td>27.308473</td>\n",
       "      <td>19.089219</td>\n",
       "      <td>18.519512</td>\n",
       "      <td>20.292008</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2Yr x 10Yr</th>\n",
       "      <td>50.914573</td>\n",
       "      <td>30.821070</td>\n",
       "      <td>22.685149</td>\n",
       "      <td>20.703677</td>\n",
       "      <td>21.210147</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5Yr x 2Yr</th>\n",
       "      <td>58.659040</td>\n",
       "      <td>35.398144</td>\n",
       "      <td>26.615354</td>\n",
       "      <td>23.334478</td>\n",
       "      <td>22.692058</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5Yr x 20Yr</th>\n",
       "      <td>39.819371</td>\n",
       "      <td>26.414806</td>\n",
       "      <td>20.719175</td>\n",
       "      <td>18.680983</td>\n",
       "      <td>18.415331</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10Yr x 5Yr</th>\n",
       "      <td>40.788047</td>\n",
       "      <td>27.525920</td>\n",
       "      <td>21.844759</td>\n",
       "      <td>19.709819</td>\n",
       "      <td>19.279674</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10Yr x 30Yr</th>\n",
       "      <td>39.968144</td>\n",
       "      <td>25.624990</td>\n",
       "      <td>19.708825</td>\n",
       "      <td>17.560324</td>\n",
       "      <td>17.228791</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20Yr x 10Yr</th>\n",
       "      <td>38.828229</td>\n",
       "      <td>24.506873</td>\n",
       "      <td>18.898226</td>\n",
       "      <td>16.580851</td>\n",
       "      <td>15.858820</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30Yr x 2Yr</th>\n",
       "      <td>35.402863</td>\n",
       "      <td>22.255319</td>\n",
       "      <td>18.380604</td>\n",
       "      <td>16.291563</td>\n",
       "      <td>14.922371</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30Yr x 20Yr</th>\n",
       "      <td>42.810531</td>\n",
       "      <td>24.695706</td>\n",
       "      <td>18.953614</td>\n",
       "      <td>16.273739</td>\n",
       "      <td>15.023706</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               -200bps    -100bps        ATM     100bps     200bps\n",
       "3Mo x 2Yr    69.649544  34.689795  16.708668  17.124383  21.387780\n",
       "3Mo x 20Yr   58.027905  31.340723  16.915073  21.086362  26.966016\n",
       "1Yr x 5Yr    67.079706  35.281748  25.181680  20.055387  16.904566\n",
       "1Yr x 30Yr   46.351442  27.308473  19.089219  18.519512  20.292008\n",
       "2Yr x 10Yr   50.914573  30.821070  22.685149  20.703677  21.210147\n",
       "5Yr x 2Yr    58.659040  35.398144  26.615354  23.334478  22.692058\n",
       "5Yr x 20Yr   39.819371  26.414806  20.719175  18.680983  18.415331\n",
       "10Yr x 5Yr   40.788047  27.525920  21.844759  19.709819  19.279674\n",
       "10Yr x 30Yr  39.968144  25.624990  19.708825  17.560324  17.228791\n",
       "20Yr x 10Yr  38.828229  24.506873  18.898226  16.580851  15.858820\n",
       "30Yr x 2Yr   35.402863  22.255319  18.380604  16.291563  14.922371\n",
       "30Yr x 20Yr  42.810531  24.695706  18.953614  16.273739  15.023706"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Show calibrated contract Black vols\n",
    "ccon_vol_lst = pd.DataFrame(ccon_vol,\n",
    "                            columns=bps_range[1:],\n",
    "                            index=ext)\n",
    "ccon_vol_lst.loc[::3,['-200bps','-100bps','ATM','100bps','200bps']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1080 with 34 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#plotting\n",
    "width = 2\n",
    "length = 1.5\n",
    "expiry_lst = np.array([0.25, 1, 2, 5, 10, 20, 30]) \n",
    "tenor_lst = np.array([2, 5, 10, 20, 30])\n",
    "bps_chg = d_ir.columns.values[1:10]\n",
    "bps_chg[4] = 0\n",
    "\n",
    "# Get the upper and lower bound of the axis for plots\n",
    "# max, min row axis label of contract vol\n",
    "max_con_vol, min_con_vol = [], []\n",
    "for i in range(len_ir):\n",
    "    max_con_vol.append(max(ccon_vol_lst.iloc[i]))\n",
    "    min_con_vol.append(min(ccon_vol_lst.iloc[i]))\n",
    "# max, min row axis label of vol\n",
    "max_q_vol, min_q_vol = [], []\n",
    "for i in range(len_ir):\n",
    "    max_q_vol.append(max(d_ir.iloc[i][1:10]))\n",
    "    min_q_vol.append(min(d_ir.iloc[i][1:10]))\n",
    "\n",
    "axis_max, axis_min = [], [] \n",
    "for i in range(len_ir):\n",
    "    axis_max.append(max(max_con_vol[i],max_q_vol[i])*1.1)\n",
    "    axis_min.append(min(min_con_vol[i],min_q_vol[i])*0.9)\n",
    "\n",
    "#Start plotting\n",
    "fig = plt.figure(figsize = (width*10,length*10))\n",
    "fig.suptitle('Implied Volatility (%) vs. BPS Change',fontsize=24)\n",
    "\n",
    "for i1 in range(len(expiry_lst)):\n",
    "    for i2 in range(len(tenor_lst)):\n",
    "        if (i1==len(expiry_lst)-1) & (i2==len(tenor_lst)-1):\n",
    "            break\n",
    "        else:\n",
    "            ind = i1*5+i2\n",
    "            ax = fig.add_subplot(len(expiry_lst),len(tenor_lst),ind+1) #plot index, plot column\n",
    "            ax.plot(bps_chg, ccon_vol[ind], 'gs')\n",
    "            ax.plot(bps_chg, d_ir.iloc[ind].values[1:10], 'r--')\n",
    "            ax.spines['top'].set_color('none')\n",
    "            ax.spines['right'].set_color('none')\n",
    "            ax.grid(linestyle='--',alpha = 0.1)\n",
    "            ax.set_ylim(axis_min[ind],axis_max[ind])\n",
    "            if i2 == 0:\n",
    "                if i1 == 0:\n",
    "                    ax.legend(['Calibrated Contract Vol','Market Quote Vol'])\n",
    "                    ax.set_ylabel('Expiry = %sMo'%(int(expiry_lst[i1]*12)), fontsize = length * 7.5)\n",
    "                else:\n",
    "                    ax.set_ylabel('Expiry = %sYr'%(int(expiry_lst[i1])), fontsize = length * 7.5)\n",
    "            if i1 == 0:\n",
    "                ax.set_title('Tenor = %sYr'%tenor_lst[i2], fontsize = width * 7.5)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Use calibrated SABR to price Swaptions with SMM"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Denote the par swap rate for the $[T_n,T_N]$ swap as $S_{n,N}$:\n",
    "$$\n",
    "S_{n,N}(t) = \\frac{D_n(t) - D_N(t)}{\\sum_{i=n+1}^{N}\\Delta_{i-1}D_i(t)}\n",
    "$$\n",
    "The term in the denominator is also called the present value of a basis point (PVBP):\n",
    "$$\n",
    "P_{n+1,N}(t) = \\sum_{i=n+1}^{N}\\Delta_{i-1}D_i(t)\n",
    "$$\n",
    "Note that a one period swap rate $S_{i,i+1}$ is equal to the LIBOR rate. We can now write the value of a payer and receiver swap as:\n",
    "$$\n",
    "V_{n,N}^{Payer} = P_{n+1,N}(t)(S_{n,N}(t)- K)\n",
    "$$\n",
    "\n",
    "$$\n",
    "V_{n,N}^{Receiver} = P_{n+1,N}(t)(K-S_{n,N}(t))\n",
    "$$\n",
    "\n",
    "If we use $P_{n+1,N}(t)$ as a numeraire, then under the measure $Q^{n+1,N}$ associated to the numeraire PVBP. All $P_{n+1,N}$ rebased values must be martingales in an arbitrage-free world.\n",
    "\n",
    "$$\n",
    "V^{payer}_{n,N}(0) = P_{n+1,N}(0)E^{n+1,N}[(S_{n,N}(T) - K)^+]\n",
    "$$\n",
    "\n",
    "Assuming $S_{n,N}$ is a lognormal martingale under $Q^{n+1,N}$, we can write down the process with Swap Market Model (SMM):\n",
    "\n",
    "$$\n",
    "dS_{n,N}(t) = \\sigma_{n,N}S_{n,N}(t)dW^{n+1,N}\n",
    "$$\n",
    "\n",
    "Using PVBP, the payoff of a payer swaption is given by:\n",
    "\n",
    "$$\n",
    "V^{payer}_{n,N}(0) = P_{n+1,N}(0)E^{n+1,N}[(S_{n,N}(T) - K)^+]\n",
    "$$\n",
    "\n",
    "$$\n",
    "=P_{n+1,N}(0)[S_{n,N}(0)\\Phi(d_1) - K\\Phi(d_2)]\n",
    "$$\n",
    "\n",
    "$$\n",
    "=P_{n+1,N}(0)Black_{76}(K,S_{n,N}(0),T,\\sigma)\n",
    "$$\n",
    "where\n",
    "$$\n",
    "d_1 = \\frac{log\\frac{S_{n,N}(0)}{K}+\\frac{1}{2}\\sigma_{n,N}^2T}{\\sigma_{n,N}\\sqrt T},  d_2 = d_1 - \\sigma_{n,N}\\sqrt T\n",
    "$$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Use calibrated SABR to price Swaptions\n",
    "from scipy.stats import norm\n",
    "def Black76(F, K, T, \n",
    "            D, # Discount factor \n",
    "            sigma, \n",
    "            option_type #'c' for call and 'p' for put\n",
    "           ):\n",
    "    c_p = option_type.lower()\n",
    "    d1 = (np.log(F/K)+(sigma**2/2)*T)/(sigma*T**0.5)\n",
    "    d2 = d1 - sigma*T**0.5\n",
    "    if c_p == 'c':\n",
    "        v = F*norm.cdf(d1)- K*norm.cdf(d2)\n",
    "    else:\n",
    "        v = K*norm.cdf(-d2) - F*norm.cdf(-d1)\n",
    "    return v*D\n",
    "\n",
    "# Here we assume a list of strike from 1% to 8%\n",
    "Fs = [0.01,0.02,0.03,0.04,0.05,0.06,0.07,0.08]\n",
    "\n",
    "# Swaption PVBP, P_{n+1,N}(t)\n",
    "# Here the PVBP is SA payment using 30/360 daycount\n",
    "def Swaption_PVBP(expiry, tenor):\n",
    "    dt = 0.5\n",
    "    start_con = int(expiry*4)\n",
    "    end_con = int((expiry+tenor)*4)\n",
    "    odf = np.array(odfs3m[start_con:end_con:2].copy())\n",
    "    return odf.sum()\n",
    "\n",
    "# Price Swaption with SABR Vol\n",
    "def Price_Swaption_SABR(expiry, tenor, # float\n",
    "                        K, # Provided Swaption Strike \n",
    "                        s_t # swaption type, P/R\n",
    "                       ):\n",
    "    if expiry==30 & tenor==30:\n",
    "        return math.nan\n",
    "    # Get Sabr Vol of the contract\n",
    "    # Get the index of the contract\n",
    "    def Find_index(arr,\n",
    "                   item):\n",
    "        ret = None\n",
    "        for i in range(len(arr)):\n",
    "            if arr[i] == item:\n",
    "                ret = i\n",
    "                break\n",
    "        return ret\n",
    "    exp_i = np.where(expiry_lst==expiry)[0][0]\n",
    "    ten_i = np.where(tenor_lst==tenor)[0][0]\n",
    "    ind = exp_i*5+ten_i\n",
    "    # Get Swaption Price\n",
    "    F = firs[ind]\n",
    "    # Vol\n",
    "    ps = para_lst[ind] # calibrated model parameters\n",
    "    con_svol = SABR(F,K,tenor,ps[0],ps[1],ps[2],ps[3]) # Get SABR vol\n",
    "    # Discount Factor\n",
    "    df = odfs3m[int(expiry*4)-1]\n",
    "    # Swaption Type\n",
    "    # Payer pay fixed, receive floating\n",
    "    if s_t == 'payer':\n",
    "        c_p = 'c'\n",
    "    else:\n",
    "        c_p = 'p'\n",
    "    PVBP = Swaption_PVBP(expiry, tenor)\n",
    "    #print('TxE=%sx%s'%(tenor,expiry),PVBP, Black76(F,K,tenor,df,con_svol/100,c_p))\n",
    "    return PVBP*Black76(F,K,tenor,df,con_svol/100,c_p)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "scrolled": false
   },
   "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>0.01</th>\n",
       "      <th>0.02</th>\n",
       "      <th>0.03</th>\n",
       "      <th>0.04</th>\n",
       "      <th>0.05</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Expiry x Tenor</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>3Mo x 2Yr</th>\n",
       "      <td>0.069165</td>\n",
       "      <td>0.032815</td>\n",
       "      <td>0.005032</td>\n",
       "      <td>0.001037</td>\n",
       "      <td>0.000476</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3Mo x 20Yr</th>\n",
       "      <td>0.758778</td>\n",
       "      <td>0.501632</td>\n",
       "      <td>0.238666</td>\n",
       "      <td>0.240571</td>\n",
       "      <td>0.276351</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1Yr x 5Yr</th>\n",
       "      <td>0.174191</td>\n",
       "      <td>0.098413</td>\n",
       "      <td>0.043427</td>\n",
       "      <td>0.013387</td>\n",
       "      <td>0.002601</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1Yr x 30Yr</th>\n",
       "      <td>0.949054</td>\n",
       "      <td>0.662307</td>\n",
       "      <td>0.413307</td>\n",
       "      <td>0.322445</td>\n",
       "      <td>0.300335</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2Yr x 10Yr</th>\n",
       "      <td>0.340665</td>\n",
       "      <td>0.213780</td>\n",
       "      <td>0.115964</td>\n",
       "      <td>0.066123</td>\n",
       "      <td>0.044645</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5Yr x 2Yr</th>\n",
       "      <td>0.050437</td>\n",
       "      <td>0.024541</td>\n",
       "      <td>0.007537</td>\n",
       "      <td>0.001583</td>\n",
       "      <td>0.000330</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5Yr x 20Yr</th>\n",
       "      <td>0.531494</td>\n",
       "      <td>0.364128</td>\n",
       "      <td>0.230863</td>\n",
       "      <td>0.149568</td>\n",
       "      <td>0.106903</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10Yr x 5Yr</th>\n",
       "      <td>0.113707</td>\n",
       "      <td>0.067642</td>\n",
       "      <td>0.032496</td>\n",
       "      <td>0.013270</td>\n",
       "      <td>0.005466</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10Yr x 30Yr</th>\n",
       "      <td>0.548016</td>\n",
       "      <td>0.387831</td>\n",
       "      <td>0.256592</td>\n",
       "      <td>0.175979</td>\n",
       "      <td>0.133380</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20Yr x 10Yr</th>\n",
       "      <td>0.109562</td>\n",
       "      <td>0.064857</td>\n",
       "      <td>0.032048</td>\n",
       "      <td>0.014160</td>\n",
       "      <td>0.006431</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30Yr x 2Yr</th>\n",
       "      <td>0.009921</td>\n",
       "      <td>0.003507</td>\n",
       "      <td>0.000442</td>\n",
       "      <td>0.000100</td>\n",
       "      <td>0.000100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30Yr x 20Yr</th>\n",
       "      <td>0.104612</td>\n",
       "      <td>0.064849</td>\n",
       "      <td>0.036262</td>\n",
       "      <td>0.019198</td>\n",
       "      <td>0.010327</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                    0.01      0.02      0.03      0.04      0.05\n",
       "Expiry x Tenor                                                  \n",
       "3Mo x 2Yr       0.069165  0.032815  0.005032  0.001037  0.000476\n",
       "3Mo x 20Yr      0.758778  0.501632  0.238666  0.240571  0.276351\n",
       "1Yr x 5Yr       0.174191  0.098413  0.043427  0.013387  0.002601\n",
       "1Yr x 30Yr      0.949054  0.662307  0.413307  0.322445  0.300335\n",
       "2Yr x 10Yr      0.340665  0.213780  0.115964  0.066123  0.044645\n",
       "5Yr x 2Yr       0.050437  0.024541  0.007537  0.001583  0.000330\n",
       "5Yr x 20Yr      0.531494  0.364128  0.230863  0.149568  0.106903\n",
       "10Yr x 5Yr      0.113707  0.067642  0.032496  0.013270  0.005466\n",
       "10Yr x 30Yr     0.548016  0.387831  0.256592  0.175979  0.133380\n",
       "20Yr x 10Yr     0.109562  0.064857  0.032048  0.014160  0.006431\n",
       "30Yr x 2Yr      0.009921  0.003507  0.000442  0.000100  0.000100\n",
       "30Yr x 20Yr     0.104612  0.064849  0.036262  0.019198  0.010327"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Calculate payer swaptions\n",
    "cs_payers = []\n",
    "for i1 in range(len_ir): # i1 is contract index\n",
    "    price_lst = []\n",
    "    for i2 in range(len(Fs)): # i2 is Forward index\n",
    "        price_lst.append(max(0.0001,Price_Swaption_SABR(exp[i1],ten[i1],Fs[i2],'payer')))\n",
    "    cs_payers.append(price_lst)\n",
    "    \n",
    "cs_payer_df = pd.DataFrame(data=cs_payers,\n",
    "             index=d_ir['Expiry x Tenor'],\n",
    "             columns=Fs)\n",
    "cs_payer_df.iloc[::3,:5:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "scrolled": false
   },
   "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>0.02</th>\n",
       "      <th>0.04</th>\n",
       "      <th>0.06</th>\n",
       "      <th>0.08</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Expiry x Tenor</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3Mo x 2Yr</th>\n",
       "      <td>0.004615</td>\n",
       "      <td>0.049819</td>\n",
       "      <td>0.126032</td>\n",
       "      <td>0.202854</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3Mo x 20Yr</th>\n",
       "      <td>0.235666</td>\n",
       "      <td>0.578513</td>\n",
       "      <td>1.247188</td>\n",
       "      <td>1.892882</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1Yr x 5Yr</th>\n",
       "      <td>0.030862</td>\n",
       "      <td>0.122859</td>\n",
       "      <td>0.286789</td>\n",
       "      <td>0.463520</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1Yr x 30Yr</th>\n",
       "      <td>0.330559</td>\n",
       "      <td>0.756312</td>\n",
       "      <td>1.491489</td>\n",
       "      <td>2.248121</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2Yr x 10Yr</th>\n",
       "      <td>0.077536</td>\n",
       "      <td>0.243968</td>\n",
       "      <td>0.525448</td>\n",
       "      <td>0.827965</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5Yr x 2Yr</th>\n",
       "      <td>0.004737</td>\n",
       "      <td>0.041031</td>\n",
       "      <td>0.098783</td>\n",
       "      <td>0.157963</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5Yr x 20Yr</th>\n",
       "      <td>0.152945</td>\n",
       "      <td>0.400968</td>\n",
       "      <td>0.797070</td>\n",
       "      <td>1.234395</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10Yr x 5Yr</th>\n",
       "      <td>0.011962</td>\n",
       "      <td>0.064444</td>\n",
       "      <td>0.160504</td>\n",
       "      <td>0.265547</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10Yr x 30Yr</th>\n",
       "      <td>0.188284</td>\n",
       "      <td>0.438244</td>\n",
       "      <td>0.833198</td>\n",
       "      <td>1.268193</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20Yr x 10Yr</th>\n",
       "      <td>0.019241</td>\n",
       "      <td>0.079052</td>\n",
       "      <td>0.178589</td>\n",
       "      <td>0.286904</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30Yr x 2Yr</th>\n",
       "      <td>0.000646</td>\n",
       "      <td>0.011256</td>\n",
       "      <td>0.025342</td>\n",
       "      <td>0.039443</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30Yr x 20Yr</th>\n",
       "      <td>0.033423</td>\n",
       "      <td>0.100932</td>\n",
       "      <td>0.200766</td>\n",
       "      <td>0.310325</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                    0.02      0.04      0.06      0.08\n",
       "Expiry x Tenor                                        \n",
       "3Mo x 2Yr       0.004615  0.049819  0.126032  0.202854\n",
       "3Mo x 20Yr      0.235666  0.578513  1.247188  1.892882\n",
       "1Yr x 5Yr       0.030862  0.122859  0.286789  0.463520\n",
       "1Yr x 30Yr      0.330559  0.756312  1.491489  2.248121\n",
       "2Yr x 10Yr      0.077536  0.243968  0.525448  0.827965\n",
       "5Yr x 2Yr       0.004737  0.041031  0.098783  0.157963\n",
       "5Yr x 20Yr      0.152945  0.400968  0.797070  1.234395\n",
       "10Yr x 5Yr      0.011962  0.064444  0.160504  0.265547\n",
       "10Yr x 30Yr     0.188284  0.438244  0.833198  1.268193\n",
       "20Yr x 10Yr     0.019241  0.079052  0.178589  0.286904\n",
       "30Yr x 2Yr      0.000646  0.011256  0.025342  0.039443\n",
       "30Yr x 20Yr     0.033423  0.100932  0.200766  0.310325"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Calculate receiver swaptions\n",
    "cs_receivers = []\n",
    "for i1 in range(len_ir): # i1 is contract index\n",
    "    price_lst = []\n",
    "    for i2 in range(len(Fs)): # i2 is Forward index\n",
    "        price_lst.append(max(0.0001,Price_Swaption_SABR(exp[i1],ten[i1],Fs[i2],'receiver')))\n",
    "    cs_receivers.append(price_lst)\n",
    "    \n",
    "cs_receivers_df  = pd.DataFrame(data=cs_receivers,\n",
    "             index=d_ir['Expiry x Tenor'],\n",
    "             columns=Fs)\n",
    "cs_receivers_df.iloc[::3,1::2] "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Part III: CMS Convexity Correction"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "scrolled": true
   },
   "source": [
    "## Model-free pricing of CMS swaps, caps and floors "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### **1. The Reference Swap Rate.**\n",
    "\n",
    "The value of the CMS swap, cap or floor is just the sum of the values of each payment. \n",
    "Any margin payments $m$ can also be valued easily. Let $R_s$ denote the reference swap rate for the CMS contract:\n",
    "$$\n",
    "\\begin{array}{c}\n",
    "[R_s - K]^+ \\cr\n",
    "[K - R_s]^+ \\cr\n",
    "\\end{array}\n",
    "$$\n",
    "\n",
    "paid at time $t_i$. The discounted value of this payment at time $t$ is: \n",
    "\n",
    "$$\n",
    "\\begin{array}{c}\n",
    "[R_s - K]^+D_(t,t_i) \\cr\n",
    "[K - R_s]^+D_(t,t_i)\\\\\\\n",
    "\\end{array}\n",
    "$$\n",
    "where $D(t,t_i)$ is the discount factor from $t$ to $t_i$.\n",
    "\n",
    "Here the contract is a standard swap that starts from $t_n$ and ends $N$ years later at $t_N$. Let $K$ denote the rate of the fixed leg. The fixed leg of this swap has a present value of:\n",
    "$$\n",
    "K\\sum_{i=n+1}^{N}\\Delta_{i-1}D_i(t_n)\\\\\\\n",
    "$$\n",
    "where $i = n+1, n+2, ..., N$, and $D_i(t_n)$ stands for $D(t_n,t_i)$\n",
    "\n",
    "$\\sum_{i=n+1}^{N}\\Delta_{i-1}D_i(t_n)$ or $P_{n+1,N}(t_n)$ is commonly known as **PVBP or PV01**.\n",
    "\n",
    "Hence, the value of the payer swap (pay fixed receive floating) can be expressed as:\n",
    "\n",
    "$$\n",
    "V^{payer}_{n,N} = [R_s- K] P_{n+1,N}(t_n)\\\\\\\n",
    "$$\n",
    "\n",
    "The floating leg is worth 1 paid at the start date $t_n$ and 1 received at the end date $t_N$. At any date $t$ the value of the payer swap is:\n",
    "$$\n",
    "V^{payer}_{n,N}(t) = D(t,t_n) - D(t,t_N) - K\\sum_{i=n+1}^{N}\\Delta_{i-1}D_i(t_n)\n",
    "$$\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### **2. Valuation.** \n",
    "\n",
    "Let $R_s$ denote the reference swap rate for the CMS contract.\n",
    "\n",
    "The payoff of a CMS caplet is $[R_s - K]^+$, paid at $t_p$\n",
    "\n",
    "On the swap's reset date/fixing date $t$, the par swap rate $R_s$ is reset and the payoff is known at $t$ as $D(t,t_p)[R_s - K]^+$, since the payment is made at $t_p$.\n",
    "\n",
    "\n",
    "Use PVBP to rebase the CMS caplet that starts at $t_n$ and ends at $t_N$, we have :\n",
    "$$\\begin{array}{c}\n",
    "\\frac{V^{CMS}_{cap}(t)}{P_{n+1,N}(t)} = \\mathrm{E}^{n+1,N}\\bigg[\\frac{V^{CMS}_{cap}(T)}{P_{n+1,N}(T)}\\bigg]\\cr\n",
    "V^{CMS}_{cap}(t) = P_{n+1,N}(t)\\mathrm{E}^{n+1,N}\\bigg[\\frac{[R_s(T)-K]^+D(T,t_p)}{P_{n+1,N}(T)}\\bigg] \\cr\n",
    "\\end{array}\n",
    "$$\n",
    "In particular, the present value of the CMS caplet can be written as:\n",
    "$$\n",
    "V^{CMS}_{cap}(0) = P_{n+1,N}(0)\\mathrm{E}^{n+1,N}\\Bigg[\\frac{[R_s(T)-K]^+D(T,t_p)}{P_{n+1,N}(T)}\\Bigg]\n",
    "$$\n",
    "\n",
    "To complete the pricing, one has to invoke a mathmetical model (Black76, Heston's, SABR model) for how $R_s(T)$ is distributed around its mean value $R_s(0)$. For example, in Black76 Model and Absolute Vol model, the swap rates are distributed respectively according to:\n",
    "$$\\begin{array}{c}\n",
    "R_s(T) = R_s(0)\\exp(\\sigma x\\sqrt{T}-\\frac{1}{2}\\sigma^2T),\\cr\n",
    "R_s(T) = R_s(0) + ax\\sqrt{T}\\cr\n",
    "\\end{array}\n",
    "$$\n",
    "where x is a standard normal random variable.\n",
    "However, these valuation process are dependent on models and are open to model risks."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### **2.1 Pricing CMS Caplets.** \n",
    "\n",
    "According to the theory of arbitrage free pricing, we can choose **any freely tradable instrument** as _numeraire_. For example, $P_{n+1,N}(t)$ is a common numeraire that can be seen as a portfolio of zero-coupon bonds. CMS are PVBP settled products. Hence, the payoff of a CMS payer contract at time T should be:\n",
    "$$\n",
    "[R_s(T) - K]^+\\\\\n",
    "$$\n",
    "_Note: In the USD market, swap-settled swaptions are liquid, participants agree on the value of PVBP.\n",
    "In the UK and Europe market, market participants use IRR to discount the cash flow, where $D(0,T) = \\frac{1}{(1+r)^T}$. Observed swap rate $S_{n,N}(T)$ can be used as a good approximation for $r$._\n",
    "\n",
    "Today's value of a CMS payer contract is:\n",
    "$$\n",
    "V^{CMS}_{cap}(0) = P_{n+1,N}(0)\\mathrm{E}^{n+1,N}\\Bigg[\\frac{[R_s-K]^+D(T,t_p)}{P_{n+1,N}(T)}\\Bigg]\n",
    "$$\n",
    "where the ratio $\\frac{D(T,t_p)}{P_{n+1,N}(T)}$is Martingale, whose expected value is today's value:\n",
    "$$\n",
    "E^{n+1,N}\\Bigg[\\frac{D(T,t_p)}{P_{n+1,N}(T)}\\Bigg] = \\frac{D(0,t_p)}{P_{n+1,N}(0)}\n",
    "$$\n",
    "\n",
    "By dividing $\\frac{D(T,t_p)}{P_{n+1,N}(T)}$ by its mean, we obtain:\n",
    "$$\n",
    "V^{CMS}_{cap}(0) = P_{n+1,N}(0)E^{n+1,N}\\Bigg[[R_s(T) - K]^+ \\frac{D(T,t_p)/P_{n+1,N}(T)}{D(0,t_p)/P_{n+1,N}(0)}\\Bigg]\n",
    "$$\n",
    "In this step, $\\frac{D(T,t_p)/P_{n+1,N}(T)}{D(0,t_p)/P_{n+1,N}(0)}$ is the Radon-Nikodym derivative used to link two different measure.\n",
    "\n",
    "\n",
    "The formula also can be written more evocatively as:\n",
    "$$\n",
    "\\begin{array}{c}\n",
    "V^{CMS}_{cap}(0) = D(0,t_p)E^{n+1,N}\\Bigg[[R_s(T) - K]^+\\Bigg]+D(0,t_p)E^{n+1,N}\\Bigg[[R_s(T) - K]^+(\\frac{D(T,t_p)/P_{n+1,N}(T)}{D(0,t_p)/P_{n+1,N}(0)}-1)\\Bigg]\n",
    "\\end{array}\n",
    "$$\n",
    "<br>\n",
    "The first term is exactly the price of a European swaption with notional $D(0,t_p)/P_{n+1,N}(0)$, regardless of which model is used to model $R_s(T)$.\n",
    "\n",
    "The last term is the \"convexity correction\". Since $R_s(T)$ is a Martingale and $[D(T,t_p)/P_{n+1,N}(T)]/[D(0,t_p)/P_{n+1,N}(0)]-1$ has an expectation of zero.\n",
    "\n",
    "The convexity correction term goes to zero linearly with the variance of the swap rate $R_s(T)$, and is much smaller than the notional price.\n",
    "\n",
    "There are **two** steps in evaluating the convexity correction. The **first** step is to _model_ the yield curve movements in a way that allows us to re-write PVBP $P_{n+1,N}$ and the zero coupon bond D(T,t_p) in terms of the reference swap rate $R_s$.\n",
    "\n",
    "Let function $G(R_s)$ denote the Martingale, so we have $D(T,t_p)/P_{n+1,N}(T)=G(R_s(T))$,$D(0,t_p)/P_{n+1,N}(0) = G(R_s(0))$. The convexity correction is then just the expected value:\n",
    "$$\n",
    "cc = D(0,t_p)E^{n+1,N}\\Bigg[[R_s(T) - K]^+(\\frac{G(R_s(T))}{G(R_s(0))}-1)\\Bigg]\n",
    "$$\n",
    "The **second** step is to evaluate this expected value.\n",
    "\n",
    "Here, we start with the street-standard model for expressing $P_{n+1,N}$ and $D(T,t_p)$ in terms of the swap rate $R_s$. This model uses bond math to obtain:\n",
    "\n",
    "$$\n",
    "G(R_s) = \\frac{R_s}{(1+R_s/q)^{\\frac{1}{q}}}\\frac{1}{1-\\frac{1}{(1+R_s/q)^n}}.\n",
    "$$\n",
    "\n",
    "We can carry out the second step by replicating the payoff of the CMS contracts as payer swaptions. For any smooth function $f(R_s)$ with $f(F) = 0$ ($F$ is the forward price under risk-neutral density), we can write the model-independent replicating formula of caplet as:\n",
    "\n",
    "$$\n",
    "h'(F)[R_s-F]^{+} + \\int_{F}^{\\infty}h''(K)[R_s-K]^{+}dK = \\left\\{\n",
    "    \\begin{array}{ll}\n",
    "        h(R_s)  & \\mbox{for  } R_s > K\\\\\n",
    "        0 & \\mbox{for  } R_s < K\n",
    "    \\end{array}\n",
    "\\right.\n",
    "$$\n",
    "Choosing\n",
    "$$\n",
    "h(R_s(T)) = \\frac{g(R_s(T))}{P_{n+1,N}(0)} = [R_s(T)-K]\\bigg(\\frac{G(R_s(T))}{G(R_s(0))}-1)\\bigg)\n",
    "$$\n",
    "\n",
    "Here, $g(x)$ is the present value of the product payoff $g(R_s(T)) = D(0,T)[R_s(T) - K]$. $h(x)$ is a function of the rebased payoff of the product. In this case, the payoff of the caplet is $[R_s-K]^+$ and the numeraire is $D(0,T)/P_{n+1,N}(T)$\n",
    "\n",
    "Sustituting into previous formula, we find that\n",
    "$$\n",
    "cc = D(0,t_p)\\Bigg[h'(F)E^{n+1,N}\\Big[R_s(T)-F\\Big]^+ + \\int_{\\infty}^{F}h''(K)E^{n+1,N}\\Big[R_s(T)-K]^+\\Big] \\Bigg]\n",
    "$$\n",
    "\n",
    "Together with the first term, the value of the caplet is:\n",
    "$$\n",
    "V^{CMS}_{cap}(0) = \\frac{D(0,t_p)}{P_{n+1,N}(0)}\\bigg[[1+h'(F)]C(F)+\\int_{F}^{\\infty}h''(K)C(K)dK\\bigg]\n",
    "$$\n",
    "where $C(x)$ is the value of the payer swaption with strike $K$ as of today: $C(K) = P_{n+1,N}(0)D(0,T)[R_s(T)-K]$\n",
    "\n",
    "\n",
    "_This formula replicates the value of the CMS caplet in terms of **European** swaptions at different strikes x. At this point some pricing systems break the integral up into 10bp or so buckets. These swaptions are then consolidated with the other European swaptions centered in each bucket. These swaptions are then consolidated with the other European swaptions in the vanilla book, and priced in the vanilla pricing system._\n",
    "\n",
    "_This \"replication method\" is the most accurate method of evaluating CMS legs. It also has the advantage of automatically making the CMS pricing and hedging consistent with the desk's handling of the rest of its vanilla book. In particular, it incorporates the desk's smile/skew corrections into the CMS pricing._"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2.2 Pricing floorlets and swaplets\n",
    "\n",
    "Repeating the arguments, the value of a CMS floorlet can be replicated by vanilla recevier swaptions in the same way:\n",
    "\n",
    "$$\n",
    "V_{floor}^{CMS}(0) = \\frac{D(0,t_p)}{P_{n+1,N}(0)}\\bigg[[1+f'(F)]P(F) - \\int_{-\\infty}^{F}f''(K)P(K)dK\\bigg]\n",
    "$$\n",
    "where\n",
    "\n",
    "$$P(K) = P_{n+1,N}(0)E^{n+1,N}\\bigg[[K-R_s(T)]^+\\bigg]\n",
    "$$\n",
    "\n",
    "is the value of the receiver swaption with strike x. Similarly, we can value a single CMS swap payment as:\n",
    "\n",
    "$$V^{CMS}_{swap}(0) = D(0,t_p)R_s(0)+h'(F)[C(F)-P(F)]+\\frac{D(0,t_p)}{P_{n+1,N}(0)}\\Bigg[\\int_{F}^{\\infty}h''(K)C(K)dK+\\int^{F}_{-\\infty}h''(K)P(K)dK\\Bigg]\n",
    "$$\n",
    "\n",
    "where $F$ is the forward price (in risk-neutral density it's $F=e^{rT}$), and $C(F) = P(F)$.\n",
    "\n",
    "and $h(K)$ is the rebased product payoff that strikes at $K$:\n",
    "\n",
    "$$h(K) = [R_s(T) - K]^+\\bigg(\\frac{G(R_s(T))}{G(R_s(0))}-1\\bigg)\n",
    "$$\n",
    "\n",
    "Unsurprisingly, the CMS caplets and floorlets follows put-call parity:\n",
    "\n",
    "$$[R_s(T)-K]+ - [K-R_s(T)]+ = R_s(T) - K\n",
    "$$\n",
    "The difference between the payoff of caplet and floorlet equals the difference between the spot rate and strike price.\n",
    "\n",
    "\n",
    "$$V^{CMS}_{cap}(t) - V^{CMS}_{floor}(t) = V^{CMS}_{swap}(t) - KD(t,t_p)\n",
    "$$\n",
    "In particular:\n",
    "\n",
    "$$V^{CMS}_{cap}(0) - V^{CMS}_{floor}(0) = V^{CMS}_{swap}(0) - KD(0,t_p)\n",
    "$$\n",
    "\n",
    "Accordingly, we can price an in-the-money CMS caplet or floorlet as a CMS swap plus an out-of-the-money floorlet or caplet.\n",
    "***\n",
    "Reference: \n",
    "Hagan P S. Convexity conundrums: Pricing CMS swaps, caps, and floors[J]. The Best of Wilmott, 2005: 305. [Hagan 2005][1].\n",
    "\n",
    "[1]: https://scholar.google.com.sg/scholar?cluster=11736429882437304210&hl=zh-CN&as_sdt=0,5       \"Hagan 2005\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### ** *Static Replication and Breeden-Litzenberger Formula**\n",
    "\n",
    "The Breeden-Litzenberger Formula allows us to use the riks-neutral density backed-out from vanilla option prices to statically replicate any twice deifferentiable European payoff $h(S_T)$. We can observe the vanilla European option prices (at the money, deep in/out of the money) from the market.\n",
    "\n",
    "Let $F = S_0e^{rT}$, we have\n",
    "$$\n",
    "V_0 = e^{-rT}E\\bigg[h(S_T)\\bigg] = \\int_{0}^{F}h(K)\\frac{\\partial^2P(K)}{\\partial{K^2}}dK + \\int_{F}^{\\infty}h(K)\\frac{\\partial^2C(K)}{\\partial{K^2}}dK\n",
    "$$\n",
    "Using integration-by-parts twice, we obtain\n",
    "$$\n",
    "V_0 = e^{-rT}h(F) + h'(F)(C(F)-P(F)) \n",
    "+\\int_{0}^{F}h''(K)P(K)dK + \\int_{F}^{\\infty}h''(K)C(K)dK\n",
    "$$\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}