{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Hedging using Machine Learning Techniques"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Summary"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As systematic and macro factors dominate the investment landscape, we see equity investors move away from one-size-fits all hedging strategies to more precise ways to separate intended and unintended risks and isolate alpha. \n",
    "\n",
    "If you are a fundamental equity investor, take concentrated positions in single names, or have significant idiosyncratic risk that is otherwise difficult to hedge, this highly customizable correlation-based approach is for you. Traditional factor based hedges can fare poorly when most of the risk cannot be explained by factors – in those instances this approach can allow for much better correlation to offset risk as well as a number of ways to guide which names get included while ensuring a high level of tradability by controlling liquidity and borrow costs.\n",
    "\n",
    "In this notebook, we will showcase how to leverage this approach through one of our most popular tools, the [marquee performance hedger](https://marquee.gs.com/s/hedging/performance).\n",
    "\n",
    "Additionally, based on feedback we have received from top users, we are adding the ability to easily run and compare hedges in python through [gs quant](https://developer.gs.com/docs/gsquant/) as well as new modeling techniques, two of whose key advantages are:\n",
    "\n",
    "* **Increased accuracy** through reduced overfitting\n",
    "* **More control** by allowing the user to specify how concentrated or diversified the hedge portfolio is\n",
    "\n",
    "\n",
    "The contents of this notebook are as follows:\n",
    "* [1 - Let's get started with gs quant](#1---Let's-get-started-with-gs-quant)\n",
    "* [2 - Calculate a hedge](#2---Calculate-a-hedge)\n",
    "* [3 - Increased accuracy](#3---Increased-accuracy)\n",
    "* [4 - More control](#4---More-control)\n",
    "* [5 - Customizing your optimization](#5---Customizing-your-optimization)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1 - Let's get started with gs quant\n",
    "Start every session with authenticating with your unique client id and secret. If you don't have a registered app, create one [here](https://marquee.gs.com/s/developer/myapps/register). `run_analytics` scope is required for the functionality covered in this example."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from gs_quant.session import GsSession, Environment\n",
    "GsSession.use(client_id=None, client_secret=None, scopes=('run_analytics',))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Below, you can set the logging level of the notebook. By default, we set the level to INFO to show informative statements about the hedging functions."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import logging\n",
    "\n",
    "logging.basicConfig(level=logging.INFO)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2 - Calculate a Hedge"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's start by calculating a hedge for Amazon with some starting parameters.\n",
    "\n",
    "To leverage the (machine learning-based) techniques in the enhanced performance hedger, please ensure:\n",
    "* `use_machine_learning` parameter is true\n",
    "\n",
    "Note: the hedge result utilizes optimal values found (using grid search) for the ML parameters, which are known as Concentration (`lasso_weight`) and Diversity (`ridge_weight`)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>country</th>\n",
       "      <th>name</th>\n",
       "      <th>transactionCost</th>\n",
       "      <th>sector</th>\n",
       "      <th>shares</th>\n",
       "      <th>assetId</th>\n",
       "      <th>bbid</th>\n",
       "      <th>currency</th>\n",
       "      <th>industry</th>\n",
       "      <th>marginalCost</th>\n",
       "      <th>advPercentage</th>\n",
       "      <th>price</th>\n",
       "      <th>weight</th>\n",
       "      <th>borrowCost</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>United States</td>\n",
       "      <td>Adobe Inc</td>\n",
       "      <td>20.263329</td>\n",
       "      <td>Information Technology</td>\n",
       "      <td>-832.032329</td>\n",
       "      <td>MA4B66MW5E27U9XPV7X</td>\n",
       "      <td>ADBE UW</td>\n",
       "      <td>USD</td>\n",
       "      <td>Software</td>\n",
       "      <td>0.261326</td>\n",
       "      <td>0.000149</td>\n",
       "      <td>310.00</td>\n",
       "      <td>-0.012897</td>\n",
       "      <td>40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>United States</td>\n",
       "      <td>Autodesk Inc</td>\n",
       "      <td>20.628566</td>\n",
       "      <td>Information Technology</td>\n",
       "      <td>-1387.449548</td>\n",
       "      <td>MA4B66MW5E27U9XPVBT</td>\n",
       "      <td>ADSK UW</td>\n",
       "      <td>USD</td>\n",
       "      <td>Software</td>\n",
       "      <td>0.214601</td>\n",
       "      <td>0.000393</td>\n",
       "      <td>149.96</td>\n",
       "      <td>-0.010403</td>\n",
       "      <td>40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>United States</td>\n",
       "      <td>Akamai Technologies Inc</td>\n",
       "      <td>26.810420</td>\n",
       "      <td>Information Technology</td>\n",
       "      <td>-4678.311848</td>\n",
       "      <td>MA4B66MW5E27U9XPVVM</td>\n",
       "      <td>AKAM UW</td>\n",
       "      <td>USD</td>\n",
       "      <td>IT Services</td>\n",
       "      <td>0.564298</td>\n",
       "      <td>0.002052</td>\n",
       "      <td>89.98</td>\n",
       "      <td>-0.021048</td>\n",
       "      <td>40</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         country                     name  transactionCost  \\\n",
       "0  United States                Adobe Inc        20.263329   \n",
       "1  United States             Autodesk Inc        20.628566   \n",
       "2  United States  Akamai Technologies Inc        26.810420   \n",
       "\n",
       "                   sector       shares              assetId     bbid currency  \\\n",
       "0  Information Technology  -832.032329  MA4B66MW5E27U9XPV7X  ADBE UW      USD   \n",
       "1  Information Technology -1387.449548  MA4B66MW5E27U9XPVBT  ADSK UW      USD   \n",
       "2  Information Technology -4678.311848  MA4B66MW5E27U9XPVVM  AKAM UW      USD   \n",
       "\n",
       "      industry  marginalCost  advPercentage   price    weight  borrowCost  \n",
       "0     Software      0.261326       0.000149  310.00 -0.012897          40  \n",
       "1     Software      0.214601       0.000393  149.96 -0.010403          40  \n",
       "2  IT Services      0.564298       0.002052   89.98 -0.021048          40  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from gs_quant.api.gs.hedges import GsHedgeApi\n",
    "from gs_quant.markets.securities import SecurityMaster, AssetIdentifier\n",
    "from datetime import date\n",
    "import pandas as pd\n",
    "\n",
    "target_asset = SecurityMaster.get_asset('AMZN UW', AssetIdentifier.BLOOMBERG_ID).get_marquee_id()\n",
    "universe = SecurityMaster.get_asset('SPX', AssetIdentifier.BLOOMBERG_ID).get_marquee_id()\n",
    "hedge_query_example = GsHedgeApi.construct_performance_hedge_query(hedge_target=target_asset, \n",
    "                                                                   universe=(universe, ),\n",
    "                                                                   notional=10e6,\n",
    "                                                                   observation_start_date=date(2019, 3, 25), \n",
    "                                                                   observation_end_date=date(2020, 3, 24), \n",
    "                                                                   backtest_start_date=date(2020, 3, 25), \n",
    "                                                                   backtest_end_date=date(2020, 4, 24),\n",
    "                                                                   use_machine_learning=True)\n",
    "\n",
    "# Calculate the hedge using the hedge_query_example as input\n",
    "results = GsHedgeApi.calculate_hedge(hedge_query_example)\n",
    "pd.DataFrame(results['result']['hedgedTarget']['constituents']).head(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3 - Increased Accuracy"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now let's compare results with and without using the new machine learning parameters and examine the first advantage - improved accuracy.\n",
    "\n",
    "We can do this by looking at the out-of-sample differences in cumulative returns of the two hedges over a hedge backtest period. Note here that we fit the hedge to the observation window and use the backtest window to see how both hedges perform. \n",
    "\n",
    "Note for this example: the ML parameters, Concentration (lasso_weight) and Diversity (ridge_weight), are being manually passed in, which is an alternative way to run the new performance hedger (compared to running grid search to find the optimal pair)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "observation_start = date(2019, 3, 25)\n",
    "observation_end = date(2020, 3, 24)\n",
    "backtest_start = observation_end\n",
    "backtest_end = date(2020, 4, 24)\n",
    "\n",
    "standard_hedger_query = GsHedgeApi.construct_performance_hedge_query(hedge_target=target_asset, \n",
    "                                                                     universe=(universe, ), \n",
    "                                                                     notional=10e6,\n",
    "                                                                     observation_start_date=observation_start, \n",
    "                                                                     observation_end_date=observation_end, \n",
    "                                                                     backtest_start_date=backtest_start, \n",
    "                                                                     backtest_end_date=backtest_end,\n",
    "                                                                     max_return_deviation=20)\n",
    "\n",
    "new_hedger_query = GsHedgeApi.construct_performance_hedge_query(hedge_target=target_asset, \n",
    "                                                                universe=(universe, ), \n",
    "                                                                notional=10e6,\n",
    "                                                                observation_start_date=observation_start,\n",
    "                                                                observation_end_date=observation_end,\n",
    "                                                                backtest_start_date=backtest_start,\n",
    "                                                                backtest_end_date=backtest_end,\n",
    "                                                                use_machine_learning=True,\n",
    "                                                                lasso_weight=5.0,\n",
    "                                                                ridge_weight=5.0,\n",
    "                                                                max_return_deviation=20)\n",
    "\n",
    "standard_results = GsHedgeApi.calculate_hedge(standard_hedger_query)\n",
    "new_hedger_results = GsHedgeApi.calculate_hedge(new_hedger_query)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "def compare_backtests_against_target_asset(new_results, standard_results, figsize=(10, 6)):\n",
    "    dates, target_returns = zip(*new_results['result']['target']['backtestPerformance'])\n",
    "    _, new_hedge_returns = zip(*new_results['result']['hedge']['backtestPerformance'])\n",
    "    _, standard_hedge_returns = zip(*standard_results['result']['hedge']['backtestPerformance'])\n",
    "    \n",
    "    results = pd.DataFrame([pd.Series(target_returns, dates, name='Target'), \n",
    "                           pd.Series(new_hedge_returns, dates, name='New Hedger'),\n",
    "                           pd.Series(standard_hedge_returns, dates, name='Standard Hedger')]).T\n",
    "    \n",
    "    results.plot(figsize=(10, 6))\n",
    "    plt.legend(fontsize=18)\n",
    "    plt.xlabel('Backtest Period', size=13)\n",
    "    plt.ylabel('Cumulative Returns (% change)', size=13)\n",
    "    plt.title('Hedge Performance Against Asset', size=22)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, we plot the cumulative returns of the target asset against the cumulative returns of each hedge."
   ]
  },
  {
   "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": [
    "compare_backtests_against_target_asset(new_hedger_results, standard_results)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As you can see above, the New Performance Hedger more closely tracks the target asset in the backtest, or \"simulated future\", period (and thus is more accurate)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 4 - More Control"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, let's demonstrate the second advantage, namely the enhanced control we have over the performance hedger results.\n",
    "\n",
    "Below, you can plot the effects of varying Concentration and Diversity on your hedge query from section 2 above. \n",
    "\n",
    "In this example, if the values for Concentration are 'None' and the values for Diversity are [10, 20], then the plotting function would run the hedge query passed in for Diversity values of 10% and 20% and then plot the weight/number of asset distributions on the y-axis and x-axis, respectively. \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# For now, use ONLY Concentration or Diversity when plotting to view the effects of this hyperparameter value changing on a hedge (set the hyperparam you aren't \n",
    "# using to None)\n",
    "hyperparams = {'Concentration': None, 'Diversity': [10, 20]}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from gs_quant.markets.hedge import Hedge\n",
    "\n",
    "# Plot results of the hedge - with emphasis on the effects that either the Concentration or Diversity hyperparameter has on the hedge\n",
    "hedge_plot = Hedge.plot_weights_against_number_of_assets(hedge_query_example, hyperparams, figsize=(10, 6))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As you can see above, varying the value of a single hyperparameter has significant effects on the weights & total number of assets of the hedge portfolio that is constructed to hedge the single target asset.\n",
    "\n",
    "For this particular example, the effect is that the weights become more balanced the more the \"Diversity\" hyperparameter increases."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 5 - Customizing your Optimization"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Since we know how to run a basic hedge, let's experiment with the concentration and diversity metrics to find an optimal hedge.\n",
    "\n",
    "Note, the hedger is flexible and allows you to choose which metric to optimize - for example, you may want to run to optimize correlation (in which case you would maximize r-squared) or to optimize transaction costs (in which case you would minimize total transaction costs)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Modify hyperparameter grid to the values you want to use to find the optimal hedge . As mentioned before, the terminology used for Concentration and Diversity\n",
    "# seen below relates to the terminology used for each term in the machine learning literature.\n",
    "# Concentration = Lasso (as a percentage)\n",
    "# Diversity = Ridge (as a percentage)\n",
    "hyperparams = {'Concentration': [0, 20, 40], 'Diversity': [10, 20]}\n",
    "\n",
    "# Modify this to optimize a metric (maximize or minimize depending on the metric. \n",
    "# See the create_optimization_mappings function for how metrics are optimized\n",
    "metric_to_optimize = 'rSquared'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's now run the optimization through the custom grid of concentration/diversity values we specified, and optimize for the specified metric. In this case, we will look for the combination of concentration and diversity values that maximizes correlation, or r-squared."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:gs_quant.markets.hedge:We are trying to maximize rSquared and will return the optimized hedge & metric value...\n",
      "INFO:gs_quant.markets.hedge:Current Hedge is using the following values for Concentration/Diversity: (0, 10)\n",
      "INFO:gs_quant.markets.hedge:Current Hedge value for rSquared: 67.6%\n",
      "INFO:gs_quant.markets.hedge:Current Hedge is using the following values for Concentration/Diversity: (0, 20)\n",
      "INFO:gs_quant.markets.hedge:Current Hedge value for rSquared: 65.3%\n",
      "INFO:gs_quant.markets.hedge:Current Hedge is using the following values for Concentration/Diversity: (20, 10)\n",
      "INFO:gs_quant.markets.hedge:Current Hedge value for rSquared: 65.6%\n",
      "INFO:gs_quant.markets.hedge:Current Hedge is using the following values for Concentration/Diversity: (20, 20)\n",
      "INFO:gs_quant.markets.hedge:Current Hedge value for rSquared: 63.4%\n",
      "INFO:gs_quant.markets.hedge:Current Hedge is using the following values for Concentration/Diversity: (40, 10)\n",
      "INFO:gs_quant.markets.hedge:Current Hedge value for rSquared: 65.0%\n",
      "INFO:gs_quant.markets.hedge:Current Hedge is using the following values for Concentration/Diversity: (40, 20)\n",
      "INFO:gs_quant.markets.hedge:Current Hedge value for rSquared: 63.1%\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The optimal pair of hyperparameters was (0, 10), achieving a value for rSquared of 67.6% during the out of sample period.\n"
     ]
    }
   ],
   "source": [
    "from gs_quant.markets.hedge import Hedge\n",
    "\n",
    "opt_hedge, opt_metric_val, opt_hyperparams = Hedge.find_optimal_hedge(hedge_query_example, hyperparams, metric_to_optimize)\n",
    "print(f'The optimal pair of hyperparameters was {opt_hyperparams}, achieving a value for {metric_to_optimize} '\n",
    "                     f'of {opt_metric_val*100:.3}% during the out of sample period.')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Stay tuned for additional ways to use the New Performance Hedger using `gs-quant`. For any questions/comments, feel free to reach out to the email distro **gs-data-ml**!"
   ]
  }
 ],
 "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": "3.7.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}