Factor_Modeling_Talk.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "<center>\n",
    "    <h1> Factor Modeling</h1>\n",
    "    \n",
    "    <h2></h2>\n",
    "        <img src=\"./images/long_short_equity_toy_example.png\" alt=\"Cross-sectional equity modeling\" style=\"width: 400px;\"/>\n",
    "    \n",
    "    <h4>Max Margenot</h4>\n",
    "    <h4>Lead - Data Science at Quantopian</h4>\n",
    "</center>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "*This presentation is for informational purposes only and does not constitute an offer to sell, a solicitation to buy, or a recommendation for any security; nor does it constitute an offer to provide investment advisory or other services by Quantopian, Inc. (\"Quantopian\"). Nothing contained herein constitutes investment advice or offers any opinion with respect to the suitability of any security, and any views expressed herein should not be taken as advice to buy, sell, or hold any security or as an endorsement of any security or company. In preparing the information contained herein, Quantopian, Inc. has not taken into account the investment needs, objectives, and financial circumstances of any particular investor. Any views expressed and data illustrated herein were prepared based upon information, believed to be reliable, available to Quantopian, Inc. at the time of publication. Quantopian makes no guarantees as to their accuracy or completeness. All information is subject to change and may quickly become unreliable for various reasons, including changes in market conditions or economic circumstances.*"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# Background\n",
    "\n",
    "* Me"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# What is a Factor?\n",
    "\n",
    "Some value assigned to a security in a universe."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "source": [
    "They come in two flavors, Alpha and Risk."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# Popularity of Factor Models\n",
    "\n",
    "\"Smart Beta\" Portfolios"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "source": [
    "* 87% of institutional investors incorporating factors into investment process\\*\n",
    "\n",
    "* 67% of into risk management process\\*\n",
    "\n",
    "* 53% of into investment strategies\\*\n",
    "\n",
    "\n",
    "\\*\"The rise of factor investing.\" BlackRock. 7 Apr. 2017"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# Goals of Factor Models\n",
    "\n",
    "Diversify with specific premia."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "source": [
    "Analyze risk."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "source": [
    "Increase returns."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# Common Factor Portfolios\n",
    "\n",
    "Most factors fall into one of five categories:\n",
    "\n",
    "* Value\n",
    "\n",
    "* Momentum\n",
    "\n",
    "* Quality\n",
    "\n",
    "* Volatility\n",
    "\n",
    "* Growth"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# Portfolio Construction\n",
    "\n",
    "Cross-Sectional Equity:\n",
    "\n",
    "* Rank all securities in your universe from lowest factor exposure to highest.\n",
    "\n",
    "* Go long on the top percentile.\n",
    "\n",
    "* Go short on the bottom percentile."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "<center>\n",
    "<img src=\"./images/long_short_equity_toy_example.png\" alt=\"Cross-sectional equity modeling\" style=\"width: 800px;\"/>\n",
    "</center>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## The Universe\n",
    "\n",
    "Sufficient liquidity."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "source": [
    "No \"hard-to-trade\" securities."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "universe = QTradableStocksUS()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "source": [
    "https://www.quantopian.com/posts/working-on-our-best-universe-yet-qtradablestocksus"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# Example Factor: Momentum"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [],
   "source": [
    "class Momentum(CustomFactor):\n",
    "    \"\"\" Momentum factor \"\"\"\n",
    "    inputs = [USEquityPricing.close,\n",
    "              Returns(window_length=126)]\n",
    "    window_length = 252\n",
    "\n",
    "    def compute(self, today, assets, out, prices, returns):\n",
    "        out[:] = ((prices[-21] - prices[-252])/prices[-252] -\n",
    "                  (prices[-1] - prices[-21])/prices[-21]) / np.nanstd(returns, axis=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# Example Factor: Momentum"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "source": [
    "<center>\n",
    "<img src=\"./images/momentum_factor_example_baskets.png\" alt=\"Mean returns by quantile for Momentum\" style=\"width: 800px;\"/>\n",
    "</center>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# Example Factor: Momentum"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "source": [
    "<center>\n",
    "<img src=\"./images/momentum_factor_example_cum_returns_quantiles.png\" alt=\"Cumulative returns by quantile for Momentum\" style=\"width: 800px;\"/>\n",
    "</center>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# Example Factor: Momentum"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "source": [
    "<center>\n",
    "<img src=\"./images/momentum_factor_example_cum_returns.png\" alt=\"Cumulative returns for momentum\" style=\"width: 800px;\"/>\n",
    "</center>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# Why Use This Method?\n",
    "\n",
    "Predicting prices is hard."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "source": [
    "With factor modeling, we capture the relative value of the long and short baskets.\n",
    "\n",
    "Key implied assumption is that the expected return of each security is proportional to its factor value."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# The Quant Workflow\n",
    "\n",
    "<center>\n",
    "<img src=\"./images/quant_workflow.png\" alt=\"Quant Equity Workflow\" style=\"width: 800px;\"/>\n",
    "</center>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# Alpha and Risk\n",
    "\n",
    "Typical form of constructing any sort of factor portfolio."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "source": [
    "How do we distinguish between Alpha and Risk?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# Portfolio Diversification\n",
    "\n",
    "Company-level risks disappear, but larger common factors remain."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "source": [
    "We can define these common factors however we like."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# Capital Asset Pricing Model\n",
    "\n",
    "Uses the market at large as a single common risk factor:\n",
    "\n",
    "$$ r_p = \\alpha + \\beta_{m} r_{m} $$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# Fama-French Factors\n",
    "\n",
    "Add additional common factors to the base CAPM:\n",
    "\n",
    "$$ r_p = \\alpha + \\beta_{m} r_{m} + \\beta_{hml} r_{hml} + \\beta_{smb} r_{smb} $$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# General Factors\n",
    "\n",
    "We can extend this as far as we want:\n",
    "\n",
    "$$ r_p = \\alpha + \\beta_0 r_0 + \\cdots + \\beta_n r_n $$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# Quantopian Risk Model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "source": [
    "<center>\n",
    "<img src=\"./images/quantopian_risk_model.png\" alt=\"Quantopian 16-factor risk model\" style=\"width: 800px;\"/>\n",
    "</center>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# Sectors"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "source": [
    "<center>\n",
    "<img src=\"./images/sector_exposures_example.png\" alt=\"Quantopian sector exposures\" style=\"width: 800px;\"/>\n",
    "</center>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# Styles"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "source": [
    "<center>\n",
    "<img src=\"./images/style_exposures_example.png\" alt=\"Quantopian style exposures\" style=\"width: 800px;\"/>\n",
    "</center>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# Performance Attribution"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "source": [
    "<center>\n",
    "<img src=\"./images/common_and_specific_example_cum_returns.png\" alt=\"Returns split into common and specific returns\" style=\"width: 800px;\"/>\n",
    "</center>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# Alpha\n",
    "\n",
    "Alpha is what remains after you have removed all common factor risk."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "source": [
    "It can only be defined in terms of what you **don't** want."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# Finding Alpha\n",
    "\n",
    "A very difficult problem."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "source": [
    "To find a reliable source of alpha, you must have a clearly defined set of common factor risks."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "source": [
    "Alpha decays."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# Tools\n",
    "\n",
    "* Alphalens: https://github.com/quantopian/alphalens\n",
    "* Pyfolio: https://github.com/quantopian/pyfolio\n",
    "* Zipline: https://github.com/quantopian/zipline\n",
    "\n",
    "# References\n",
    "\n",
    "* Quantopian Lecture Series - https://quantopian.com/lectures \n",
    "    * https://www.quantopian.com/lectures/the-capital-asset-pricing-model-and-arbitrage-pricing-theory\n",
    "    * https://www.quantopian.com/lectures/factor-analysis-with-alphalens\n",
    "    * https://www.quantopian.com/lectures/risk-constrained-portfolio-optimization\n",
    "    \n",
    "* Quantopian Risk Model - https://www.quantopian.com/papers/risk"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "<div style=\"text-align:center\">\n",
    "    <center><img src=\"./images/twitter_logo.png\" alt='Twitter' style=\"width: 50px;\"/></center>\n",
    "    <h3>@clean_utensils</h3>\n",
    "</div>\n",
    "\n",
    "<div style=\"text-align:center\">\n",
    "    <center><img src=\"./images/github_logo.png\" alt=\"Github\" style=\"width: 50px;\"/></center>\n",
    "    <h3>@mmargenot</h3>\n",
    "</div>\n",
    "\n",
    "<div style=\"text-align:center\">\n",
    "    <center><img src=\"./images/logo-q-press-logo-red.png\" alt=\"Quantopian\" style=\"width: 100px;\"/></center>\n",
    "    <h3>max@quantopian.com</h3>\n",
    "</div>\n",
    "\n",
    "<div style=\"text-align:center\">\n",
    "    <center></center>\n",
    "    <h3>Enter the Quantopian Daily Contest at www.quantopian.com/contest</h3>\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "*This presentation is for informational purposes only and does not constitute an offer to sell, a solicitation to buy, or a recommendation for any security; nor does it constitute an offer to provide investment advisory or other services by Quantopian, Inc. (\"Quantopian\"). Nothing contained herein constitutes investment advice or offers any opinion with respect to the suitability of any security, and any views expressed herein should not be taken as advice to buy, sell, or hold any security or as an endorsement of any security or company. In preparing the information contained herein, Quantopian, Inc. has not taken into account the investment needs, objectives, and financial circumstances of any particular investor. Any views expressed and data illustrated herein were prepared based upon information, believed to be reliable, available to Quantopian, Inc. at the time of publication. Quantopian makes no guarantees as to their accuracy or completeness. All information is subject to change and may quickly become unreliable for various reasons, including changes in market conditions or economic circumstances.*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[NbConvertApp] Converting notebook ./Factor_Modeling_Talk.ipynb to slides\n",
      "[NbConvertApp] Writing 279166 bytes to ./Factor_Modeling_Talk.slides.html\n"
     ]
    }
   ],
   "source": [
    "!jupyter nbconvert --to slides ./Factor_Modeling_Talk.ipynb --reveal-prefix ./reveal.js"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "celltoolbar": "Slideshow",
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}