{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Lucas Asset Pricing with advanced Approximation Methods"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### João Brogueira and Fabian Schütze"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Introduction"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This note describes why and how we modified the computer code of the original <a href=\"https://github.com/QuantEcon/QuantEcon.py/blob/master/quantecon/models/lucastree.py\">lucastree.py</a> module. We briefly reformulate Lucas' asset pricing problem as found in the  <a href=\"http://quant-econ.net/py/lucas_model.html\">lecture notes</a> . Denote by $y$ the fruit of the tree. The fruit’s growth rate follows the process $G(y,z') = y^\\alpha z'$ with $z' \\sim \\log N(0,\\sigma^2)$. The investor has CRRA preferences with curvature parameter $\\gamma$ and discount factor $\\beta$. Following <a href=\"https://www.jstor.org/stable/1913837?seq=1#page_scan_tab_contents\">Lucas (1978)</a> , the pricing function, $p(y)$, solves the functional equation:\n",
    "\n",
    "$$ f(y) = h(y) + \\beta \\int_Z f(G(y,z')) Q(dz').$$\n",
    "with\n",
    "\\begin{align*}\n",
    "   f(y) &= p(y)y^{-\\gamma}, \\\\ \n",
    "   h(y) &= \\beta \\int_Z \\left( G(y,z') \\right)^{1-\\gamma} Q(dz') = \\beta y^{ (1-\\gamma)\\alpha } \\exp \\left( (1-\\gamma)^2  \\sigma^2/2 \\right).\n",
    "\\end{align*}\n",
    "\n",
    "We want the numeric solution $f$ to comply with theoretical predictions about its functional form. In the following, it is first documented under which circumstances $h$ transmits montoncity and concavity onto  $f$. In particular, we prove that if $G$ is strictly increasing and concave <sup><a href=\"#fn1\" id=\"ref1\">1</a></sup>, $h$ transmits the sign of its first and second derivatives onto $f$. Additionally, we show that if both $G$ and $h$ are strictly decreasing and convex,  $f$ is strictly decreasing and convex as well. The solution to the functional equation is numerically obtained by iterating the contraction mapping $Tf(y) = h(y) + \\beta \\int_Z f(G(y,z')) Q(dz')$ until the distance between two successive iterations is smaller than a tolerance criteria. To compute the integral numerically,  $f(G(y,z'))$ needs to be evaluated at arguments $y$ that are not on the grid. This is a chance to impose the properties of $h$ onto $f$ through an appropriate approximation routine. This note discusses how to implement such a routine at the end.\n",
    "\n",
    "<sup id=\"fn1\">1. For the sake of brevity, when writing strictly increasing and concave we really mean strictly increasing and strictly concave. Also, strictly decreasing and convex refers to strictly decreasing and strictly convex, etc. <a href=\"#ref1\" t></a></sup>\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Theoretical Predictions about the Functional Form of the Solution to Lucas' Asset Pricing Equation "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This section documents under which circumstances $f$ inherits the sign of the first and second derivatives of $h$. In the following, suppose all necessary assumptions to guarantee a unique solution to Lucas' asset pricing problem are satisfied. One assumption is that the function $h$ is bounded in the supremum norm. Numercially, the assumption is satisfied because the  lower end of the interval $Y$ is striclty positive and because $Y$ is bounded. Theoretically, one can prove that the $h$ needs only be bounded in a weighted supremum norm when the parameter $\\alpha > 0$. Based on exercise 9.7 of the book by <a href=\"http://www.hup.harvard.edu/catalog.php?isbn=9780674750968\">Stokey and Lucas with Prescott (1989)</a>, we prove the following proposition: \n",
    "\n",
    "**Proposition** \n",
    "\n",
    "1. Suppose $G$ is strictly increasing and concave in $y$. If $h$ is sctrictly increasing and convave, $f$ is strictly increasing and concave. If $h$ is sctrictly decreasing and convex, $f$ is strictly decreasing and convex.\n",
    "2. Suppose $G$ is strictly decreasing and convex in $y$. If $h$ is strictly decreasing and convex, $f$ is strictly decreasing and convex.\n",
    "\n",
    "**Proof**\n",
    "\n",
    "**1** Following the notation of the  <a href=\"http://quant-econ.net/py/lucas_model.html\">lecture notes</a>, denote by $cb\\mathbf{R}_+$ the set of continuous and bounded functions $f:\\mathbf{R}_{+} \\rightarrow \\mathbf{R}_{+}$ . The set $cb'\\mathbf{R}_{+} \\subset cb\\mathbf{R}_{+}$ is the set of continuous, bounded, nondecreasing and concave functions, and $cb''\\mathbf{R}_{+} \\subset cb'\\mathbf{R}_{+}$ imposes additionally strict monotonicity and concavity. We want to show that the contraction operator $T$ maps any function $\\tilde{f} \\in cb'\\mathbf{R}_{+}$ into the subset $cb''\\mathbf{R}_{+}$. As the solution to the functional equation is characterized by $Tf = f$ and $cb'\\mathbf{R}_{+}$ is a closed set, if the operator $T$ transforms any nondecreasing and concave function into a strictly increasing and concave function, then $f$ is strictly increasing and concave (Corollary 1 of the Contraction Mapping Theorem in Stokey and Lucas with Prescott (1989), p. 52).\n",
    "  \n",
    "To show the desired result, suppose first that  $h$ is strictly increasing and concave and pick any $\\tilde{f} \\in cb'\\mathbf{R}_{+}$. To begin, study whether $T\\tilde{f}$ is strictly increasing. For any pair $\\hat{y},y \\in Y$ with $\\hat{y} > y$, the function $T\\tilde{f}$ satisfies: \n",
    "\n",
    "\\begin{align*}\n",
    "T\\tilde{f}(\\hat{y}) &= h(\\hat{y}) + \\beta \\int_Z \\tilde{f}( G(\\hat{y},z')) Q(dz')\\\\\n",
    "\t\t\t   \t  \t&> h(y) + \\beta \\int_Z \\tilde{f}( G(y,z')) Q(dz')\\\\\n",
    "\t\t   \t    \t&= T\\tilde{f}(y).\n",
    "\\end{align*}\n",
    "                         \n",
    "The inequality holds because $G$ and $h$ are strictly increasing and $\\tilde{f}$ is nondecreasing. Hence, $T\\tilde{f}$ is strictly increasing. \n",
    "\n",
    "To analyze concavity, define $y_{\\omega} = \\omega y + (1-\\omega) y'$, for any $y,y' \\in Y$, $y \\neq y'$, and $0 < \\omega < 1$. The strict concavity form of $h$ and $G$, together with $\\tilde{f}$ being concave, ensure that:\n",
    "\n",
    "\\begin{align*}\n",
    "T\\tilde{f}(y_\\omega) &= h(y_\\omega) + \\beta \\int_Z \\tilde{f}( G(y_\\omega,z')) Q(dz') \\\\\n",
    "\t       \t\t     &> \\omega \\left[ h(y) + \\beta \\int_Z \\tilde{f}( G(y,z')) Q(dz') \\right] + (1 - \\omega) \\left[  h(y') + \\beta \\int_Z \\tilde{f}( G(y',z')) Q(dz') \\right] \\\\\n",
    "\t\t  \t \t\t &= \\omega T\\tilde{f}(y) + (1-\\omega) T \\tilde{f}(y').\n",
    "\\end{align*}\n",
    "\n",
    "The function $T\\tilde{f}$ is stricly concave. Taken together, we know that for any $\\tilde{f} \\in cb'\\mathbf{R}_{+}$, $T\\tilde{f} \\in cb''\\mathbf{R}_{+}$. Hence, $f$ must be an element of the set $cb''\\mathbf{R}_+$, guaranteeing that $f$ has the same functional form as $h$.<br> <br>\n",
    "Now, suppose $h$ is convex and decreasing. We could again define the operator $T$ as $Tf(y) = h(y) + \\beta \\int_Z f(G(y,z'))  Q(dz')$ and study into which subset a candidate solution is mapped into. To facilitate analysis though, take a different route. Look at the modified operator \n",
    "$$Tf_{-} = h_{-} + \\beta \\int_Z f_{-} (G(y,z')) Q(z'),$$\n",
    "with $h_{-}\t = -h$ and $f_{-} = -f$. Under the same assumptions guaranteeing a unique solution to the original contraction mapping, there exists a unique solution to the modified contraction mapping. As $h_{-}$ is strictly increasing and concave, the proof above applies to the modified contraction mapping. As $f_{-}$ is strictly increasing and concave, $f$ is strictly decreasing and convex and inherits the properties of $h$. \n",
    "\n",
    "**2** As both $G$ and $h$ are strictly decreasing and convex, one can proceed in a similar fashion as in case (1.) to show that $h$ transmits its functional form to $f$. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The different cases of the proposition can be rephrased in terms of the values of the parameters $\\gamma,\\alpha$. The functional form of $h$ is jointly determined by $\\gamma,\\alpha$ as $h(y) = y^{(1-\\gamma)\\alpha} \\exp \\left( (1-\\gamma)^2 \\sigma^2/2 \\right)$. If $0 < \\alpha < 1$, $G$ is strictly increasing and concave and case (1.) of the proposition applies. If  $0 < \\gamma < 1$, $f$ is strictly increasing and concave. If $\\gamma > 1$, $f$ is strictly decreasing and convex. In contrast, suppose $-1 < \\alpha < 0$. If $0 < \\gamma < 1$ case (2.) of the proposition applies and $f$ is strictly decreasing and convex. If $\\gamma > 1$, theory does not offer any help in determining the functional form of $f$. In this situation $G$ is decreasing and convex, while $h$ is increasing. Our proposition is deliberately more restrictive than the one in exercise 9.7 of Stokey and Lucas with Prescott (1989). Because we can calculate the functions $f$ analytically for the special cases of $\\alpha \\in \\left\\lbrace 0,1\\right\\rbrace$, numercial techniques are not needed."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Imposing the functional form of $h$ onto $f$ through advanced approximation "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This section describes how we impose the functional form of $h$ onto $f$. The solution to the functional equation is numerically obtained by iterating the contraction mapping $Tf(y) = h(y) + \\beta \\int_Z f(G(y,z')) Q(dz')$ until the distance between two successive iterations is smaller than a tolerance criteria. To compute the integral numerically,  $f(G(y,z'))$ needs to be evaluated at arguments $x$ that are not on the grid through numerical approximation. This approximation is a chance to impose the properties of $h$ onto $f$. The grid points are a set $Y_{\\text{Grid}} = \\left\\lbrace y_1,y_2,\\ldots, y_{N-1},y_N \\right\\rbrace \\subset Y$, with $y_l < y_m$ if $l < m$, $l,m \\in \\mathbf{N}$. Point $x \\in Y$ is not on the gird. If $y_1 < x < y_N$ we interpolate the functional $f$ at $x$ by:  \n",
    "\n",
    "\\begin{equation}\n",
    "f(x) = f(y_L) + \\dfrac{f(y_H) - f(y_L)}{h(y_H) - h(y_L)}  \\left( h(x) - h(y_L) \\right).\n",
    "\\end{equation}\n",
    "\n",
    "with $y_L = \\max \\left\\lbrace y_i \\in Y_{\\text{Grid}} : y_i < x  \\right\\rbrace$ and $y_H = \\min \\left\\lbrace y_i \\in Y_{\\text{Grid}}: y_i > x \\right\\rbrace$. For any point $x$ lower than $y_1$ or higher than $y_N$, we define the function value as:\n",
    "\n",
    "\\begin{align}\n",
    "f(x) = \n",
    "   \\begin{cases}\n",
    "   f(y_1) + \\dfrac{f(y_1) - f(y_2)}{h(y_1) - h(y_2)} \\left(h(x) - h(y_1) \\right) & \\text{if } x < y_1,\\\\\n",
    "   f(y_N) + \\dfrac{f(y_N) - f(y_{N-1})}{h(y_N) - h(y_{N-1})} \\left( h(x) - h(y_N) \\right) & \\text{if } x > y_N.\n",
    "   \\end{cases}\n",
    "\\end{align}\n",
    "\n",
    "The approximation transmits the slope and shape of the function $h$ onto $f$ as $f'(x) \\propto h'(x)$ and $f''(x) \\propto h''(x)$ because the ratio in front of $h(x)$ is always positive. The function `interpolationFunction` of the modified  `lucastree.py` module converts this idea into computer code. The entire module is contained in the next cell."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Overwriting ./lucastree.py\n"
     ]
    }
   ],
   "source": [
    "%%writefile ./lucastree.py\n",
    "r\"\"\"\n",
    "Filename: lucastree.py\n",
    "\n",
    "Authors: Joao Brogueira and Fabian Schuetze \n",
    "\n",
    "This file is a slight modification of the lucastree.py file \n",
    "by Thomas Sargent, John Stachurski, Spencer Lyon under the \n",
    "quant-econ project. We don't claim authorship of the entire file,\n",
    "but full responsability for it and any existing mistakes.\n",
    "\n",
    "Solves the price function for the Lucas tree in a continuous state\n",
    "setting, using piecewise linear approximation for the sequence of\n",
    "candidate price functions.  The consumption endownment follows the log\n",
    "linear AR(1) process\n",
    "\n",
    ".. math::\n",
    "\n",
    "    log y' = \\alpha log y + \\sigma \\epsilon\n",
    "\n",
    "where y' is a next period y and epsilon is an iid standard normal shock.\n",
    "Hence\n",
    "\n",
    ".. math::\n",
    "\n",
    "    y' = y^{\\alpha} * \\xi,\n",
    "\n",
    "where\n",
    "\n",
    ".. math::\n",
    "\n",
    "    \\xi = e^(\\sigma * \\epsilon)\n",
    "\n",
    "The distribution phi of xi is\n",
    "\n",
    ".. math::\n",
    "\n",
    "    \\phi = LN(0, \\sigma^2),\n",
    "    \n",
    "where LN means lognormal.\n",
    "\n",
    "\"\"\"\n",
    "#from __future__ import division  # == Omit for Python 3.x == #\n",
    "import numpy as np\n",
    "from scipy.stats import lognorm\n",
    "from scipy.integrate import fixed_quad\n",
    "from quantecon.compute_fp import compute_fixed_point\n",
    "\n",
    "\n",
    "class LucasTree(object):\n",
    "\n",
    "    \"\"\"\n",
    "    Class to solve for the price of a tree in the Lucas\n",
    "    asset pricing model\n",
    "\n",
    "    Parameters\n",
    "    ----------\n",
    "    gamma : scalar(float)\n",
    "        The coefficient of risk aversion in the investor's CRRA utility\n",
    "        function\n",
    "    beta : scalar(float)\n",
    "        The investor's discount factor\n",
    "    alpha : scalar(float)\n",
    "        The correlation coefficient in the shock process\n",
    "    sigma : scalar(float)\n",
    "        The volatility of the shock process\n",
    "    grid : array_like(float), optional(default=None)\n",
    "        The grid points on which to evaluate the asset prices. Grid\n",
    "        points should be nonnegative. If None is passed, we will create\n",
    "        a reasonable one for you\n",
    "\n",
    "    Attributes\n",
    "    ----------\n",
    "    gamma, beta, alpha, sigma, grid : see Parameters\n",
    "    grid_min, grid_max, grid_size : scalar(int)\n",
    "        Properties for grid upon which prices are evaluated\n",
    "    init_h : array_like(float)\n",
    "        The functional values h(y) with grid points being arguments \n",
    "    phi : scipy.stats.lognorm\n",
    "        The distribution for the shock process\n",
    "\n",
    "    Notes\n",
    "    -----\n",
    "    This file is a slight modification of the lucastree.py file \n",
    "    by Thomas Sargent, John Stachurski, Spencer Lyon, [SSL]_ under the \n",
    "    quant-econ project. We don't claim authorship of the entire file,\n",
    "    but full responsability for it and any existing mistakes.\n",
    "\n",
    "    References\n",
    "    ----------\n",
    "    .. [SSL] Thomas Sargent, John Stachurski and Spencer Lyon, lucastree.py,\n",
    "    GitHub repository, \n",
    "    https://github.com/QuantEcon/QuantEcon.py/blob/master/quantecon/models/lucastree.py\n",
    "\n",
    "    Examples\n",
    "    --------\n",
    "    >>> tree = LucasTree(gamma=2, beta=0.95, alpha=0.90, sigma=0.1)\n",
    "    >>> grid, price_vals = tree.grid, tree.compute_lt_price()\n",
    "\n",
    "    \"\"\"\n",
    "\n",
    "    def __init__(self, gamma, beta, alpha, sigma, grid=None):\n",
    "        self.gamma = gamma\n",
    "        self.beta = beta\n",
    "        self.alpha = alpha\n",
    "        self.sigma = sigma\n",
    "\n",
    "        # == set up grid == #\n",
    "        if grid is None:\n",
    "            (self.grid, self.grid_min,\n",
    "             self.grid_max, self.grid_size) = self._new_grid()\n",
    "        else:\n",
    "            self.grid = np.asarray(grid)\n",
    "            self.grid_min = min(grid)\n",
    "            self.grid_max = max(grid)\n",
    "            self.grid_size = len(grid)\n",
    "\n",
    "        # == set up distribution for shocks == #\n",
    "        self.phi = lognorm(sigma)\n",
    "\n",
    "        # == set up integration bounds. 4 Standard deviations. Make them\n",
    "        # private attributes b/c users don't need to see them, but we\n",
    "        # only want to compute them once. == #\n",
    "        self._int_min = np.exp(-5 * sigma)\n",
    "        self._int_max = np.exp(5 * sigma)\n",
    "\n",
    "        # == Set up h for the Lucas Operator == #\n",
    "        self.init_h = self.h(self.grid)\n",
    "\n",
    "    def h(self, x):\n",
    "        \"\"\"\n",
    "        Compute the function values of h in the Lucas operator. \n",
    "\n",
    "        Parameters\n",
    "        ----------\n",
    "        x : array_like(float)\n",
    "        The arguments over which to computer the function values \n",
    "\n",
    "        Returns\n",
    "        -------\n",
    "        h : array_like(float)\n",
    "        The functional values \n",
    "\n",
    "        Notes\n",
    "        -----\n",
    "        Recall the functional form of h \n",
    "\n",
    "        .. math:: h(x) &= \\beta * \\int_Z u'(G(x,z)) phi(dz)\n",
    "                       &= \\beta x**((1-\\gamma)*\\alpha) * \\exp((1-\\gamma)**2 *\\sigma /2) \n",
    "\n",
    "        \"\"\"\n",
    "        alpha, gamma, beta, sigma = self.alpha, self.gamma, self.beta, self.sigma\n",
    "        h = beta * x**((1 - gamma) * alpha) * \\\n",
    "            np.exp((1 - gamma)**2 * sigma**2 / 2) * np.ones(x.size)\n",
    "\n",
    "        return h\n",
    "\n",
    "    def _new_grid(self):\n",
    "        \"\"\"\n",
    "        Construct the default grid for the problem\n",
    "\n",
    "        This is defined to be np.linspace(0, 10, 100) when alpha >= 1\n",
    "        and 100 evenly spaced points covering 4 standard deviations\n",
    "        when alpha < 1\n",
    "\n",
    "        \"\"\"\n",
    "        grid_size = 50\n",
    "        if abs(self.alpha) >= 1.0:\n",
    "            grid_min, grid_max = 0.1, 10\n",
    "        else:\n",
    "            # == Set the grid interval to contain most of the mass of the\n",
    "            # stationary distribution of the consumption endowment == #\n",
    "            ssd = self.sigma / np.sqrt(1 - self.alpha**2)\n",
    "            grid_min, grid_max = np.exp(-4 * ssd), np.exp(4 * ssd)\n",
    "\n",
    "        grid = np.linspace(grid_min, grid_max, grid_size)\n",
    "\n",
    "        return grid, grid_min, grid_max, grid_size\n",
    "\n",
    "    def integrate(self, g, int_min=None, int_max=None):\n",
    "        \"\"\"\n",
    "        Integrate the function g(z) * self.phi(z) from int_min to\n",
    "        int_max.\n",
    "\n",
    "        Parameters\n",
    "        ----------\n",
    "        g : function\n",
    "            The function which to integrate\n",
    "\n",
    "        int_min, int_max : scalar(float), optional\n",
    "            The bounds of integration. If either of these parameters are\n",
    "            `None` (the default), they will be set to 4 standard\n",
    "            deviations above and below the mean.\n",
    "\n",
    "        Returns\n",
    "        -------\n",
    "        result : scalar(float)\n",
    "            The result of the integration\n",
    "\n",
    "        \"\"\"\n",
    "        # == Simplify notation == #\n",
    "        phi = self.phi\n",
    "        if int_min is None:\n",
    "            int_min = self._int_min\n",
    "        if int_max is None:\n",
    "            int_max = self._int_max\n",
    "\n",
    "        # == set up integrand and integrate == #\n",
    "        integrand = lambda z: g(z) * phi.pdf(z)\n",
    "        result, error = fixed_quad(integrand, int_min, int_max, n=20)\n",
    "        return result, error\n",
    "\n",
    "    def Approximation(self, x, grid, f):\n",
    "        r\"\"\"\n",
    "        Approximates the function f at given sample points x.\n",
    "\n",
    "        Parameters\n",
    "        ----------\n",
    "        x: array_like(float)\n",
    "            Sample points over which the function f is evaluated\n",
    "\n",
    "        grid: array_like(float) \n",
    "            The grid values representing the domain of f \n",
    "\n",
    "        f: array_like(float)\n",
    "            The function values of f over the grid \n",
    "\n",
    "        Returns:\n",
    "        --------\n",
    "        fApprox: array_like(float)\n",
    "            The approximated function values at x\n",
    "\n",
    "        Notes\n",
    "        -----\n",
    "        Interpolation is done by the following function:\n",
    "\n",
    "        .. math:: f(x) = f(y_L) + \\dfrac{f(y_H) - f(y_L)}{h(y_H) - h(y_L)} (h(x) - h(y_L) ).\n",
    "\n",
    "        Extrapolation is done as follows:\n",
    "\n",
    "        .. math:: f(x) = \n",
    "        \\begin{cases}\n",
    "        f(y_1) + \\dfrac{f(y_1) - f(y_2)}{h(y_1) - h(y_2)} \\left(h(x) - h(y_1) \\right) & \\text{if } x < y_1,\\\\\n",
    "        f(y_N) + \\dfrac{f(y_N) - f(y_{N-1})}{h(y_N) - h(y_{N-1})} \\left( h(x) - h(y_N) \\right) & \\text{if } x > y_N.\n",
    "        \\end{cases}\n",
    "\n",
    "        The approximation routine imposes the functional\n",
    "        form of the function :math:`h` onto the function math:`f`, as stated\n",
    "        in chapter 9.2 (in particular theorem 9.6 and 9.7 and exercise 9.7) of the \n",
    "        book by Stokey, Lucas and Prescott (1989).\n",
    "\n",
    "        \"\"\"\n",
    "        # == Initalize and create empty arrays to be filled in the == #\n",
    "        gamma, sigma, beta = self.gamma, self.sigma, self.beta\n",
    "        hX, hGrid = self.h(x), self.init_h\n",
    "        fL, fH, fApprox = np.empty_like(x), np.empty_like(x), np.empty_like(x)\n",
    "        hL, idxL, idxH, hH = np.empty_like(x), np.empty_like(\n",
    "            x), np.empty_like(x), np.empty_like(x)\n",
    "\n",
    "        # == Create Boolean array to determine which sample points are used for interpoltion\n",
    "        # and which are used for extrapolation == #\n",
    "        lower, middle, upper = (x < grid[0]), (x > grid[0]) & (\n",
    "            x < grid[-1]), (x > grid[-1])\n",
    "\n",
    "        # == Calcualte the indices of y_L, idxL[index], and y_H ,idxH[index], that are below and above a sample point, called value.\n",
    "        # In the notation of the interpolation routine, these indices are used to pick the function values\n",
    "        # f(y_L),f(y_H),h(y_L) and h(y_H) == #\n",
    "        for index, value in enumerate(x):\n",
    "            # Calculates the indices of y_L\n",
    "            idxL[index] = (np.append(grid[grid <= value], grid[0])).argmax()\n",
    "            idxH[index] = min(idxL[index] + 1, len(grid) - 1)\n",
    "            fL[index] = f[idxL[index]]\n",
    "            fH[index] = f[idxH[index]]\n",
    "            hL[index] = hGrid[idxL[index]]\n",
    "            hH[index] = hGrid[idxH[index]]\n",
    "\n",
    "        # == Interpolation == #\n",
    "        if self.alpha != 0:\n",
    "            ratio = (fH[middle] - fL[middle]) / (hH[middle] - hL[middle])\n",
    "        elif self.alpha == 0:\n",
    "            # If self.alpha ==0, `ratio` is zero, as hH == hL\n",
    "            ratio = (hH[middle] - hL[middle])\n",
    "        fApprox[middle] = fL[middle] + ratio * (hX[middle] - hL[middle])\n",
    "\n",
    "        # == Extrapolation == #\n",
    "        if self.alpha != 0:\n",
    "            fApprox[lower] = f[\n",
    "                0] + (f[0] - f[1]) / (hGrid[0] - hGrid[1]) * (hX[lower] - hGrid[0])\n",
    "            fApprox[upper] = f[-1] + \\\n",
    "                (f[-1] - f[-2]) / (hGrid[-1] - hGrid[-2]) * \\\n",
    "                (hX[upper] - hGrid[-1])\n",
    "        elif self.alpha == 0:\n",
    "            fApprox[lower] = f[0]\n",
    "            fApprox[upper] = f[-1]\n",
    "\n",
    "        return fApprox\n",
    "\n",
    "    def lucas_operator(self, f, Tf=None):\n",
    "        \"\"\"\n",
    "        The approximate Lucas operator, which computes and returns the\n",
    "        updated function Tf on the grid points.\n",
    "\n",
    "        Parameters\n",
    "        ----------\n",
    "        f : array_like(float)\n",
    "            A candidate function on R_+ represented as points on a grid\n",
    "            and should be flat NumPy array with len(f) = len(grid)\n",
    "\n",
    "        Tf : array_like(float)\n",
    "            Optional storage array for Tf\n",
    "\n",
    "        Returns\n",
    "        -------\n",
    "        Tf : array_like(float)\n",
    "            The updated function Tf\n",
    "\n",
    "        Notes\n",
    "        -----\n",
    "        The argument `Tf` is optional, but recommended. If it is passed\n",
    "        into this function, then we do not have to allocate any memory\n",
    "        for the array here. As this function is often called many times\n",
    "        in an iterative algorithm, this can save significant computation\n",
    "        time.\n",
    "\n",
    "        \"\"\"\n",
    "        grid,  h = self.grid, self.init_h\n",
    "        alpha, beta = self.alpha, self.beta\n",
    "\n",
    "        # == set up storage if needed == #\n",
    "        if Tf is None:\n",
    "            Tf = np.empty_like(f)\n",
    "\n",
    "        # == Apply the T operator to f == #\n",
    "        Af = lambda x: self.Approximation(x, grid, f)\n",
    "\n",
    "        for i, y in enumerate(grid):\n",
    "            Tf[i] = h[i] + beta * self.integrate(lambda z: Af(y**alpha * z))[0]\n",
    "\n",
    "        return Tf\n",
    "\n",
    "    def compute_lt_price(self, error_tol=1e-7, max_iter=600, verbose=0):\n",
    "        \"\"\"\n",
    "        Compute the equilibrium price function associated with Lucas\n",
    "        tree lt\n",
    "\n",
    "        Parameters\n",
    "        ----------\n",
    "        error_tol, max_iter, verbose\n",
    "            Arguments to be passed directly to\n",
    "            `quantecon.compute_fixed_point`. See that docstring for more\n",
    "            information\n",
    "\n",
    "\n",
    "        Returns\n",
    "        -------\n",
    "        price : array_like(float)\n",
    "            The prices at the grid points in the attribute `grid` of the\n",
    "            object\n",
    "\n",
    "        \"\"\"\n",
    "        # == simplify notation == #\n",
    "        grid, grid_size = self.grid, self.grid_size\n",
    "        lucas_operator, gamma = self.lucas_operator, self.gamma\n",
    "\n",
    "        # == Create storage array for compute_fixed_point. Reduces  memory\n",
    "        # allocation and speeds code up == #\n",
    "        Tf = np.empty(grid_size)\n",
    "\n",
    "        # == Initial guess, just a vector of ones == #\n",
    "        f_init = np.ones(grid_size)\n",
    "        f = compute_fixed_point(lucas_operator, f_init, error_tol,\n",
    "                                max_iter, verbose, Tf=Tf)\n",
    "\n",
    "        price = f * grid**gamma\n",
    "\n",
    "        return price\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The following two figures plot the functions $h,f$ and their first and second differences for parameters $(\\gamma,\\alpha) \\in \\left\\lbrace (2,0.75),(0.5,0.75),(0.5,-0.75) \\right\\rbrace$. Note that the x-axis is in indices instead of grid values because the grid values change with different parameters. The graph illustrates that the  sign of the slope and shape of $h$ is transmitted to $f$. We used $|\\alpha| = 0.75$ because it generates a relatively strong visual slope of $h$. Our unit testing function also consider autoregressive parameters of $|\\alpha| \\in \\left\\lbrace 0.25, 0.5 \\right\\rbrace$. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA54AAALGCAYAAAAtJU7JAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n+v9ZkhOS68TdBAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f6a56073630>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA54AAALFCAYAAACrsTxnAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f6a55ffc6a0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "from lucastree import LucasTree\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "\n",
    "# first element gamma, second element alpha\n",
    "vector = np.array([[2, 0.75], [0.5, 0.75], [0.5, -0.75]])\n",
    "tree = LucasTree(gamma=2, beta=0.95, alpha=0.5, sigma=0.1)\n",
    "h, hdiff, hdiff2 = np.empty((len(tree.grid), vector.shape[0])), np.empty(\n",
    "    (len(tree.grid) - 1, vector.shape[0])), np.empty((len(tree.grid) - 2, vector.shape[0]))\n",
    "for idx, element in enumerate(vector):\n",
    "    tree = LucasTree(gamma=element[0], beta=0.95, alpha=element[1], sigma=0.1)\n",
    "    h[:, idx] = tree.h(tree.grid)\n",
    "    hdiff[:, idx] = np.ediff1d(h[:, idx])\n",
    "    hdiff2[:, idx] = np.ediff1d(hdiff[:, idx])\n",
    "fig1, (ax1, ax2, ax3) = plt.subplots(3, 1, sharex='col')\n",
    "annotation = ['(gamma,alpha): ' + str((i)) for i in vector]\n",
    "ax1.plot(h)\n",
    "ax2.plot(hdiff)\n",
    "ax3.plot(hdiff2)\n",
    "ax1.set_title('Plot of h')\n",
    "ax2.set_title('Plot of the first difference of h')\n",
    "ax3.set_title('Plot of the second difference of h')\n",
    "ax3.legend(annotation, loc='upper center', bbox_to_anchor=(0.5, -0.07),\n",
    "           ncol=2, fancybox=True, shadow=True, fontsize=10)\n",
    "fig1.suptitle(\n",
    "    'Plot of the function h and its first and second difference', fontsize=15)\n",
    "fig1.set_size_inches(15.5, 10.5)\n",
    "fig1.show()\n",
    "\n",
    "\n",
    "# first element gamma, second element alpha\n",
    "vector = np.array([[2, 0.75], [0.5, 0.75], [0.5, -0.75]])\n",
    "tree = LucasTree(gamma=2, beta=0.95, alpha=0.5, sigma=0.1)\n",
    "f, fdiff, fdiff2 = np.empty((len(tree.grid), vector.shape[0])), np.empty(\n",
    "    (len(tree.grid) - 1, vector.shape[0])), np.empty((len(tree.grid) - 2, vector.shape[0]))\n",
    "price = np.empty_like(f)\n",
    "for idx, element in enumerate(vector):\n",
    "    tree = LucasTree(gamma=element[0], beta=0.95, alpha=element[1], sigma=0.1)\n",
    "    price[:, idx], grid = tree.compute_lt_price(), tree.grid\n",
    "    f[:, idx] = price[:, idx] * grid**(-element[0])\n",
    "    fdiff[:, idx] = np.ediff1d(f[:, idx])\n",
    "    fdiff2[:, idx] = np.ediff1d(fdiff[:, idx])\n",
    "fig2, (ax1, ax2, ax3) = plt.subplots(3, 1, sharex='col')\n",
    "annotation = ['(gamma,alpha): ' + str((i)) for i in vector]\n",
    "ax1.plot(f)\n",
    "ax2.plot(fdiff)\n",
    "ax3.plot(fdiff2)\n",
    "ax1.set_title('Plot of f')\n",
    "ax2.set_title('Plot of the first difference of f')\n",
    "ax3.set_title('Plot of the second difference of f')\n",
    "ax3.legend(annotation, loc='upper center', bbox_to_anchor=(0.5, -0.05),\n",
    "           ncol=2, fancybox=True, shadow=True, fontsize=11)\n",
    "fig2.suptitle(\n",
    "    'Plot of the function f and its first and second difference', fontsize=15)\n",
    "fig2.set_size_inches(15.5, 10.5)\n",
    "fig2.show()\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The two following figures graphs the solution $f$ if $\\alpha \\in \\left\\lbrace 0,1 \\right\\rbrace$. If  dividends follow an i.i.d. process, ($\\alpha = 0$) the function $f$ is constant. We reproduce the numerical results in the top panel of the following figure. The lower subplot graphs the price dividend ratio when dividend growth follows and i.i.d process ($\\alpha =1$). As predicted by theory, the price dividend ratio is a constant. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA5EAAAKiCAYAAABPS1gIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f6a55deecc0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "from lucastree import LucasTree\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "beta, gamma, sigma = 0.95, 2, 0.1\n",
    "\n",
    "tree = LucasTree(gamma=gamma, beta=beta, alpha=1, sigma=sigma)\n",
    "priceLinear, grid = tree.compute_lt_price(), tree.grid\n",
    "fig1, (ax1, ax2) = plt.subplots(2, 1)\n",
    "theoreticalPDRatio = np.ones(len(grid)) * beta * np.exp((1 - gamma) **\n",
    "                                                        2 * sigma**2 / 2) / (1 - beta * np.exp((1 - gamma)**2 * sigma**2 / 2))\n",
    "ax1.plot(grid, priceLinear / grid, grid, theoreticalPDRatio, 'g^')\n",
    "annotation = ['Numerical Solution', 'Analytical Solution']\n",
    "ax1.legend(annotation)\n",
    "ax1.set_title('price dividend ratio for alpha = 1')\n",
    "ax1.set_ylim([min(priceLinear / grid) - 1, max(priceLinear / grid) + 1])\n",
    "tree = LucasTree(gamma=gamma, beta=beta, alpha=0, sigma=sigma)\n",
    "priceFalling, grid = tree.compute_lt_price(), tree.grid\n",
    "theoreticalF = np.ones(len(grid)) * beta * \\\n",
    "    np.exp((1 - gamma)**2 * sigma**2 / 2) / (1 - beta)\n",
    "f = priceFalling * grid**(-2)\n",
    "ax2.plot(grid, f, grid, theoreticalF, 'g^')\n",
    "ax2.set_ylim([min(f) - 1, max(f) + 1])\n",
    "annotation = ['Numerical Solution', 'Analytical Solution']\n",
    "ax2.legend(annotation)\n",
    "ax2.set_title('function f for alpha = 0')\n",
    "fig1.set_size_inches(15.5, 10.5)\n",
    "fig1.suptitle(\n",
    "    'Plot of the function f  and the price dividend ratio for alpha=0 and alpha=1 respecitvely', fontsize=15)\n",
    "fig1.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Finally, we report the unit testing function accompanying our lucastree.py module. This file tests if the functional form of $f$ adheres to the theoretical predicitons as outlined by the Proposition above. The file can be run from the Shell.  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Overwriting ./test_lucastree.py\n"
     ]
    }
   ],
   "source": [
    "%%writefile ./test_lucastree.py\n",
    "\"\"\"\n",
    "filename: test_lucastree.py\n",
    "\n",
    "Authors: Joao Brogueira and Fabian Schuetze \n",
    "\n",
    "This file contains for different test for the \n",
    "lucastree.py file \n",
    "\n",
    "Functions\n",
    "---------\n",
    "    compute_lt_price()      [Status: Tested in test_ConstantPDRatio, test_ConstantF,\n",
    "                            test_slope_f, test_shape_f]\n",
    "\n",
    "\"\"\"\n",
    "\n",
    "import unittest\n",
    "from lucastree import LucasTree  # This relative importing doesn't work!\n",
    "import numpy as np\n",
    "\n",
    "\n",
    "class Testlucastree(unittest.TestCase):\n",
    "\n",
    "    \"\"\"\n",
    "    Test Suite for lucastree.py based on the outout of the \n",
    "    LucasTree.compute_lt_price() function.\n",
    "\n",
    "    \"\"\"\n",
    "    # == Parameter values applicable to all test cases == #\n",
    "    beta = 0.95\n",
    "    sigma = 0.1\n",
    "\n",
    "    # == Paramter values for different tests == #\n",
    "    ConstantPD = np.array([2, 1])\n",
    "    ConstantF = np.array([2, 0])\n",
    "    FunctionalForm = np.array([[2, 0.75], [2, 0.5], [2, 0.25], [0.5, 0.75], [\n",
    "                              0.5, 0.5], [0.5, 0.25], [0.5, -0.75], [0.5, -0.5], [0.5, -0.25]])\n",
    "\n",
    "    # == Tolerance Criteria == #\n",
    "    Tol = 1e-2\n",
    "\n",
    "    def setUp(self):\n",
    "        self.storage = lambda parameter0, parameter1: LucasTree(gamma=parameter0, beta=self.beta, alpha=parameter1,\n",
    "                                                                sigma=self.sigma)\n",
    "\n",
    "    def test_ConstantPDRatio(self):\n",
    "        \"\"\"\n",
    "        Test whether the numerically computed price dividend ratio is \n",
    "        identical to its theoretical counterpart when dividend \n",
    "        growth follows an idd process\n",
    "\n",
    "        \"\"\"\n",
    "        gamma, alpha = self.ConstantPD\n",
    "        tree = self.storage(gamma, alpha)\n",
    "        grid = tree.grid\n",
    "        theoreticalPDRatio = np.ones(len(grid)) * self.beta * np.exp(\n",
    "            (1 - gamma)**2 * self.sigma**2 / 2) / (1 - self.beta * np.exp((1 - gamma)**2 * self.sigma**2 / 2))\n",
    "        self.assertTrue(\n",
    "            np.allclose(theoreticalPDRatio, tree.compute_lt_price() / grid, atol=self.Tol))\n",
    "\n",
    "    def test_ConstantF(self):\n",
    "        \"\"\"\n",
    "        Tests whether the numericlaly obtained solution, math:`f` \n",
    "        to the functional equation :math:`f(y) = h(y) + \\beta \\int_Z f(G(y,z')) Q(z')`\n",
    "        is identical to its theoretical counterpart, when divideds follow an \n",
    "        iid process \n",
    "\n",
    "        \"\"\"\n",
    "        gamma, alpha = self.ConstantF\n",
    "        tree = self.storage(gamma, alpha)\n",
    "        grid = tree.grid\n",
    "        theoreticalF = np.ones(len(\n",
    "            grid)) * self.beta * np.exp((1 - gamma)**2 * self.sigma**2 / 2) / (1 - self.beta)\n",
    "        self.assertTrue(np.allclose(\n",
    "            theoreticalF, tree.compute_lt_price() * grid**(-gamma), atol=self.Tol))\n",
    "\n",
    "    def test_slope_f(self):\n",
    "        \"\"\"\n",
    "        Tests whether the first difference of the numerically obtained function \n",
    "        :math:`f` is has the same sign as the first difference of the function \n",
    "        :math:`h`.\n",
    "\n",
    "        Notes\n",
    "        -----\n",
    "        This test is motivated by Theorem 9.7 ans exercise 9.7c) of the \n",
    "        book by Stokey, Lucas and Prescott (1989)\n",
    "\n",
    "        \"\"\"\n",
    "        for parameters in self.FunctionalForm:\n",
    "            gamma, alpha = parameters\n",
    "            tree = self.storage(gamma, alpha)\n",
    "            f = tree.compute_lt_price() * tree.grid ** (-gamma)\n",
    "            h = tree.h(tree.grid)\n",
    "            fdiff, hdiff = np.ediff1d(f), np.ediff1d(h)\n",
    "            if all(hdiff > 0):\n",
    "                self.assertTrue(all(fdiff > 0))\n",
    "            elif all(hdiff < 0):\n",
    "                self.assertTrue(all(fdiff < 0))\n",
    "\n",
    "    def test_shape_f(self):\n",
    "        \"\"\"\n",
    "        Tests whether the second difference of the numerically obtained function \n",
    "        :math:`f` is has the same sign as the second difference of the function \n",
    "        :math:`h`.\n",
    "\n",
    "        Notes\n",
    "        -----\n",
    "        This test is motivated by Theorem 9.8 ans exercise 9.7d) of the \n",
    "        book by Stokey, Lucas and Prescott (1989)\n",
    "\n",
    "        \"\"\"\n",
    "        for parameters in self.FunctionalForm:\n",
    "            gamma, alpha = parameters\n",
    "            tree = self.storage(gamma, alpha)\n",
    "            f = tree.compute_lt_price() * tree.grid ** (-gamma)\n",
    "            h = tree.h(tree.grid)\n",
    "            fdiff, hdiff = np.ediff1d(f), np.ediff1d(h)\n",
    "            fdiff2, hdiff2 = np.ediff1d(fdiff), np.ediff1d(hdiff)\n",
    "            if all(hdiff2 > 0):\n",
    "                self.assertTrue(all(fdiff2 > 0))\n",
    "            elif all(hdiff2 < 0):\n",
    "                self.assertTrue(all(fdiff2 < 0))\n",
    "\n",
    "    def tearDown(self):\n",
    "        pass\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "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": "3.5.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}