var_demo.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "# -*- coding: utf-8 -*-\n",
    "\"\"\"\n",
    "Created on Apr  5 11:11:08 2019\n",
    "\n",
    "@author: timpara\n",
    "\"\"\"\n",
    "\n",
    "import quandl\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "from scipy.special import ndtri\n",
    "import scipy.stats as scs\n",
    "import statsmodels.api as sm\n",
    "import matplotlib as mpl\n",
    "import matplotlib.pyplot as plt\n",
    "import os\n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Setup"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "tickers={\"power\":\"CHRIS/ICE_DPB1\",\"coal\":\"CHRIS/ICE_ATW1\",\"carbon\":\"CHRIS/ICE_C1\",\"eu\":\"CHRIS/CME_EC1\"}\n",
    "alpha=0.05\n",
    "quandl_api_key=os.environ.get('QUANDLAPIKEY')\n",
    "#power: euro\n",
    "#coal: usd per ton\n",
    "# I think carbon emmission certificates ECX EUA Futures, are listed as CHRIS/ICE_C1\n",
    "positions=[-100,55,90]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Helper functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "def dN(x, mu, sigma):\n",
    "    ''' Probability density function of a normal random variable x.\n",
    "\n",
    "    Parameters\n",
    "    ==========\n",
    "    mu : float\n",
    "        expected value\n",
    "    sigma : float\n",
    "        standard deviation\n",
    "\n",
    "    Returns\n",
    "    =======\n",
    "    pdf : float\n",
    "        value of probability density function\n",
    "    '''\n",
    "    z = (x - mu) / sigma\n",
    "    pdf = np.exp(-0.5 * z ** 2) / np.sqrt(2 * np.pi * sigma ** 2)\n",
    "    return pdf\n",
    "\n",
    "def quotes_returns(data):\n",
    "    ''' Plots Cumulative and daily PnL. '''\n",
    "    plt.figure(figsize=(9, 6))\n",
    "    plt.subplot(211)\n",
    "    data.cumsum().plot()\n",
    "    plt.ylabel('Cumulative PnL')\n",
    "    plt.grid(True)\n",
    "    plt.axis('tight')\n",
    "\n",
    "    plt.subplot(212)\n",
    "    data.plot()\n",
    "    plt.ylabel('Daily PnL')\n",
    "    plt.grid(True)\n",
    "    plt.axis('tight')\n",
    "    plt.savefig('quotes_returns.pdf')\n",
    "# histogram of annualized daily log returns\n",
    "\n",
    "\n",
    "def return_histogram(data):\n",
    "    ''' Plots a histogram of the PnL. '''\n",
    "    plt.figure(figsize=(9, 5))\n",
    "    x = np.linspace(min(data), max(data), 100)\n",
    "    plt.hist(np.array(data), bins=50, normed=True)\n",
    "    y = dN(x, np.mean(data), np.std(data))\n",
    "    plt.plot(x, y, linewidth=2)\n",
    "    plt.xlabel('PnL')\n",
    "    plt.ylabel('frequency/probability')\n",
    "    plt.grid(True)\n",
    "    plt.savefig('return_histogram.pdf')\n",
    "\n",
    "# Q-Q plot of annualized daily log returns\n",
    "\n",
    "\n",
    "def return_qqplot(data):\n",
    "    ''' Generates a Q-Q plot of the returns.'''\n",
    "    plt.figure(figsize=(9, 5))\n",
    "    sm.qqplot(data, line='s')\n",
    "    plt.grid(True)\n",
    "    plt.xlabel('theoretical quantiles')\n",
    "    plt.ylabel('sample quantiles')\n",
    "    plt.savefig(\"return_qq_plot.pdf\")\n",
    "def print_statistics(data):\n",
    "    print(\"RETURN SAMPLE STATISTICS\")\n",
    "    print(\"---------------------------------------------\")\n",
    "    print(\"Mean of Daily  Log Returns %9.6f\" % np.mean(data))\n",
    "    print(\"Mean of Annua. Log Returns %9.6f\" %\n",
    "          (np.mean(data) * 252))\n",
    "    print(\"Std  of Annua. Log Returns %9.6f\" %\n",
    "          (np.std(data) * np.sqrt(252)))\n",
    "    print(\"---------------------------------------------\")\n",
    "    print(\"Skew of Sample Log Returns %9.6f\" % scs.skew(data))\n",
    "    print(\"Skew Normal Test p-value   %9.6f\" %\n",
    "          scs.skewtest(data)[1])\n",
    "    print(\"---------------------------------------------\")\n",
    "    print(\"Kurt of Sample Log Returns %9.6f\" % scs.kurtosis(data))\n",
    "    print(\"Kurt Normal Test p-value   %9.6f\" %\n",
    "          scs.kurtosistest(data)[1])\n",
    "    print(\"---------------------------------------------\")\n",
    "    print(\"Normal Test p-value        %9.6f\" %\n",
    "          scs.normaltest(data)[1])\n",
    "    print(\"---------------------------------------------\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Main functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "def getData(tickers,quandl_api_key):\n",
    "    '''\n",
    "    Download data from quandl. Converte USD/Coal to Euro/Coal if Coal and Eu are given as ticker.\n",
    "    \n",
    "    Input:\n",
    "    tickers: dictionary, Quandl tickers\n",
    "    \n",
    "    Output:\n",
    "    data: levels, float\n",
    "    returns: log returns based on settle price\n",
    "    '''\n",
    "    data=pd.DataFrame()\n",
    "    for key,value in tickers.items():\n",
    "        data[str(key+\"_settle\")]=quandl.get(value,start_date=\"2018-01-01\",api_key=quandl_api_key)[\"Settle\"]\n",
    "    if \"coal\" and \"eu\" in tickers.keys():\n",
    "        data[\"coal_settle\"]=data[\"coal_settle\"]*data[\"eu_settle\"]\n",
    "        data.drop(columns=\"eu_settle\",inplace=True)\n",
    "    #data[\"power_settle\"]=data[\"power_settle\"]*-1\n",
    "    data.fillna(method='ffill', inplace=True)\n",
    "    returns=np.log(data/data.shift(1)).dropna()\n",
    "    max_record=len(returns)\n",
    "    \n",
    "    return data, returns,max_record\n",
    "    \n",
    "    #data_quandl.drop(columns=[\"eu_settle\"],inplace=True)\n",
    "def getPortfolioStatistics(returns,data,positions):\n",
    "    portfolio={}\n",
    "    portfolio[\"capital_per_position\"]=np.array([data.power_settle[0]*positions[0],\n",
    "                                data.coal_settle[0]*positions[1],\n",
    "                               data.carbon_settle[0]*positions[2]])\n",
    "    #total euro exposure\n",
    "    portfolio[\"capital_total\"]=np.sum(portfolio[\"capital_per_position\"])\n",
    "    \n",
    "    #weights\n",
    "    weights=portfolio[\"capital_per_position\"]/np.sum(np.abs(portfolio[\"capital_per_position\"]))\n",
    "    #covariance matrix\n",
    "    cov_mat_assets=np.cov(returns,rowvar=False)\n",
    "    #volatility assets\n",
    "    vol_assets=np.sqrt(np.diag(cov_mat_assets))\n",
    "    #portfolio return\n",
    "    ret_port=returns*portfolio[\"capital_per_position\"]\n",
    "\n",
    "\n",
    "    #portfolio cova\n",
    "    cov_port=np.matmul(cov_mat_assets,weights)\n",
    "    portfolio[\"pnl\"]=ret_port\n",
    "    #portfolio volatility\n",
    "    portfolio[\"vol_port_euro\"] = np.sqrt(np.matmul(portfolio[\"capital_per_position\"].T,np.matmul(cov_mat_assets,portfolio[\"capital_per_position\"])))\n",
    "    portfolio[\"vol_port\"] = np.sqrt(np.matmul(weights.T,np.matmul(cov_mat_assets,weights)))\n",
    "    return portfolio,weights,cov_mat_assets,ret_port,cov_port,vol_assets\n",
    "\n",
    "# Weighted Historical Simulation Approach. Defualt Significance Level is 95%\n",
    "def hist_simul(return_pad, positions, max_record, eta=.99, signi=95):\n",
    "    # Calculate Historical Daily Profits of Given Positions\n",
    "    hist_profit = return_pad * positions\n",
    "    hist_profit['total'] = hist_profit.apply(lambda x: x.sum(), axis=1)\n",
    "    # Calculate Weights of Different Dates\n",
    "    hist_weights = list(map(lambda x: eta ** (max_record - x - 1) * (1 - eta) / (1 - eta ** max_record),\n",
    "                            [x for x in range(max_record)]))\n",
    "    weighted_hist = np.column_stack((hist_profit.total.tolist(), hist_weights))\n",
    "    # Sort Profits (Low to High)\n",
    "    arg = np.argsort(weighted_hist[:, 0])\n",
    "    weighted_hist = weighted_hist[arg]\n",
    "    # Calculate Cumulative Weights\n",
    "    cum_weights = np.cumsum(weighted_hist[:, 1], axis=0)\n",
    "    weighted_hist = np.column_stack((weighted_hist, cum_weights))\n",
    "    # Find Profits Around Significance Level\n",
    "    arg = cum_weights <= (1 - signi / 100)\n",
    "    arg2 = cum_weights >= (1 - signi / 100)\n",
    "    # Interpolation\n",
    "    ip_mat = np.row_stack((weighted_hist[arg][-1, [0, 2]], weighted_hist[arg2][0, [0, 2]]))\n",
    "    hs_var = (ip_mat[1, 0] - ip_mat[0, 0]) / (ip_mat[1, 1] - ip_mat[0, 1]) * ((1 - signi / 100) - ip_mat[0, 1]) + ip_mat[0, 0]\n",
    "    return hs_var\n",
    "\n",
    "\n",
    "# Naive Historical Simulation Approach. Defualt Significance Level is 95%\n",
    "def naive_hist(return_pad, positions, signi=95):\n",
    "    hist_profit = return_pad * positions\n",
    "    hist_total = hist_profit.sum(axis=1)\n",
    "    nh_var = np.percentile(hist_total, 100 - signi)\n",
    "    return nh_var\n",
    "\n",
    "\n",
    "# Weighted Delta-Normal Simulation W or W/O RiskMetrics. Default Significance Level is 95%\n",
    "def delta_norm(return_pad, positions, max_record, lmbd=.94, signi=95, rm_sign=1):\n",
    "    # Employ RiskMetrics Weights\n",
    "    dn_weights = list(map(lambda x: (1 - lmbd) * lmbd ** (max_record - x - 1), [x for x in range(max_record)]))\n",
    "    if not rm_sign:\n",
    "        dn_weights = 1 / max_record * np.ones(max_record)\n",
    "    # Calculate Weighted/Unweighted COV, Assuming Zero-Mean\n",
    "    weighted_cov = np.dot(return_pad.T * dn_weights, return_pad)\n",
    "    delta = np.sqrt(np.dot(np.dot(positions, weighted_cov), positions))\n",
    "    dn_var = delta * ndtri(1 - signi / 100)\n",
    "    return dn_var\n",
    "\n",
    "\n",
    "# Monte Carlo Approach. Default Significance Level is 95%\n",
    "def monte_carlo(return_pad, positions, max_record, signi=95, iters=100000, lmbd=.94, rm_sign=1):\n",
    "    # Calculate COV\n",
    "    cov = return_pad.cov()\n",
    "    if rm_sign:\n",
    "        # Employ RiskMetrics Weights\n",
    "        mc_weights = list(map(lambda x: (1 - lmbd) * lmbd ** (max_record - x - 1), [x for x in range(max_record)]))\n",
    "        cov = np.dot(return_pad.T * mc_weights, return_pad)\n",
    "    # Generate Simulated returns\n",
    "    mc_simul = np.random.multivariate_normal(np.zeros(return_pad.shape[1]), cov, iters)\n",
    "    simulated_profit = mc_simul * positions\n",
    "    simulated_total = simulated_profit.sum(axis=1)\n",
    "    mc_var = np.percentile(simulated_total, 100 - signi)\n",
    "    return mc_var\n",
    "def variance_covariance_decomp(portfolio,weights,cov_mat_assets,ret_port,cov_port,vol_assets):\n",
    "    #Calculate the var in euro and percent\n",
    "    VaR_port_euro=scs.norm.ppf(1-alpha)*portfolio[\"vol_port_euro\"]\n",
    "    VaR_port=scs.norm.ppf(1-alpha)*portfolio[\"vol_port\"]\n",
    "\n",
    "    #VaR per asset\n",
    "    VaRi=scs.norm.ppf(1-alpha)*vol_assets*portfolio[\"capital_per_position\"]\n",
    "    # undiversified VaR as the sum of the VaRs of the individual assets.\n",
    "    uVar = np.sum(VaRi)\n",
    "    #beta of portfolio\n",
    "    #beta = np.divide(cov_mat_assets,np.square(portfolio[\"vol_port\"]))\n",
    "\n",
    "    # Calculate Marginal VaR\n",
    "    mVaR = scs.norm.ppf(1-alpha)*np.dot(cov_mat_assets,portfolio[\"capital_per_position\"])/portfolio[\"vol_port_euro\"]\n",
    "    mVaR_euro=mVaR * portfolio[\"capital_per_position\"]\n",
    "    # component var\n",
    "    cVaR = 100*mVaR * portfolio[\"capital_per_position\"]/VaR_port_euro\n",
    "    return cVaR,VaR_port_euro,mVaR_euro\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Run script"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "data, returns,max_record=getData(tickers)\n",
    "portfolio,weights,cov_mat_assets,ret_port,cov_port,vol_assets=getPortfolioStatistics(returns,data,positions)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.6/dist-packages/matplotlib/__init__.py:886: MatplotlibDeprecationWarning: \n",
      "examples.directory is deprecated; in the future, examples will be found relative to the 'datapath' directory.\n",
      "  \"found relative to the 'datapath' directory.\".format(key))\n",
      "/usr/local/lib/python3.6/dist-packages/matplotlib/__init__.py:886: MatplotlibDeprecationWarning: \n",
      "examples.directory is deprecated; in the future, examples will be found relative to the 'datapath' directory.\n",
      "  \"found relative to the 'datapath' directory.\".format(key))\n",
      "/usr/local/lib/python3.6/dist-packages/matplotlib/axes/_axes.py:6499: MatplotlibDeprecationWarning: \n",
      "The 'normed' kwarg was deprecated in Matplotlib 2.1 and will be removed in 3.1. Use 'density' instead.\n",
      "  alternative=\"'density'\", removal=\"3.1\")\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RETURN SAMPLE STATISTICS\n",
      "---------------------------------------------\n",
      "Mean of Daily  Log Returns -8.079405\n",
      "Mean of Annua. Log Returns -2036.010040\n",
      "Std  of Annua. Log Returns 1915.693942\n",
      "---------------------------------------------\n",
      "Skew of Sample Log Returns -0.520068\n",
      "Skew Normal Test p-value    0.000217\n",
      "---------------------------------------------\n",
      "Kurt of Sample Log Returns  5.566667\n",
      "Kurt Normal Test p-value    0.000000\n",
      "---------------------------------------------\n",
      "Normal Test p-value         0.000000\n",
      "---------------------------------------------\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 648x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 648x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 648x360 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#insert graphical staff here and some descriptive analysis\n",
    "quotes_returns(portfolio[\"pnl\"])\n",
    "return_histogram(portfolio[\"pnl\"])\n",
    "return_qqplot(portfolio[\"pnl\"])\n",
    "print_statistics(portfolio[\"pnl\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "196.97995391665296"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "-np.quantile(portfolio[\"pnl\"],0.05)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "cVaR,VaR_port_euro,mVaR_euro=variance_covariance_decomp(portfolio,weights,cov_mat_assets,ret_port,cov_port,vol_assets)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Portfolio VaR and decomposed VaR\n",
      "---------------------------------------------\n",
      "Daily VaR using Variance-Covariance Method    198.80 Euro\n",
      "1-Day VaR confidence level alpha      0.05\n",
      " Component VaR:\n",
      "Asset: power iVaR: 133.85 and 67.33 %\n",
      "Asset: coal iVaR: 67.1 and 33.75 %\n",
      "Asset: carbon iVaR: -2.15 and -1.08 %\n"
     ]
    }
   ],
   "source": [
    "print(\"Portfolio VaR and decomposed VaR\")\n",
    "print(\"---------------------------------------------\")\n",
    "print(\"Daily VaR using Variance-Covariance Method %9.2f Euro\" % np.round(VaR_port_euro,2)) \n",
    "print(\"1-Day VaR confidence level alpha %9.2f\" %alpha)\n",
    "print(\" Component VaR:\")\n",
    "k=0\n",
    "for i in tickers.keys():\n",
    "    print(\"Asset: {} iVaR: {} and {} %\".format(i,np.round(mVaR_euro[k],2),np.round(cVaR[k],2)))\n",
    "    k+=1\n",
    "    if k==3:\n",
    "        break\n",
    "        \n",
    "        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([-4810.     ,  6435.24695,   700.2    ])"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "portfolio[\"capital_per_position\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Benchmark Variance-Covariance method:\n",
      "Tomorrow VaR for the portfolio is $ 221.03 using historical simulation method.\n",
      "Tomorrow VaR for the portfolio is $ 324.25 using delta-normal method.\n",
      "Tomorrow VaR for the portfolio is $ 326.76 using Monte Carlo method.\n"
     ]
    }
   ],
   "source": [
    "print(\"Benchmark Variance-Covariance method:\")\n",
    "\n",
    "hs_var = -hist_simul(returns, portfolio[\"capital_per_position\"], max_record)\n",
    "dn_var = -delta_norm(returns, portfolio[\"capital_per_position\"], max_record)\n",
    "mc_var = -monte_carlo(returns, portfolio[\"capital_per_position\"], max_record)\n",
    "print('Tomorrow VaR for the portfolio is Euro', round(hs_var, 2), 'using historical simulation method (RiskMetrics weighting).')\n",
    "print('Tomorrow VaR for the portfolio is Euro', round(dn_var, 2), 'using delta-normal method (RiskMetrics weighting).')\n",
    "print('Tomorrow VaR for the portfolio is Euro', round(mc_var, 2), 'using Monte Carlo method.')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([-4810.     ,  6435.24695,   700.2    ])"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "portfolio[\"capital_per_position\"]"
   ]
  },
  {
   "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": "3.6.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}