ipynb/kerk_phillips-dsge_grid.ipynb

{
    "cells": [
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "# Solving and Simulating a DSGE Model in Python\n",
                "\n",
                "## Using Value-Function Iteration on a Grid\n",
                "### by [Kerk L. Phillips](https://sites.google.com/site/kerkphillips/home), July 2018\n",
                "\n",
                "The code in this Jupyter notebook was written using Python 3.6. It also uses the python function, `rouwen.py`. This file can be downloaded at the Python4econ blog - [http://python4econ.blogspot.com/2013/10/rouwenhorst-1995-markov-transition.html](http://python4econ.blogspot.com/2013/10/rouwenhorst-1995-markov-transition.html).  Alternatively, both that file and this notebook can be found in the public GitHub repository [https://github.com/kerkphil/Jupyter_Notebooks](https://github.com/kerkphil/Jupyter_Notebooks)."
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "### This notebook explains how to set up a simple DSGE model and how to solve and simulate it in Python using value-function iteration on a full cartesian grid."
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "We begin by writing down the behavioral equations for a simple DSGE model.\n",
                "\n",
                "We use the following definitions:\n",
                "\n",
                "$k_t$ is the capital stock in period $t$.\n",
                "\n",
                "$z_t$ is the percent deviation of technology from its long-run value.\n",
                "\n",
                "$r_t$ is the rental rate on capital.\n",
                "\n",
                "$w_t$ is the wage rate.\n",
                "\n",
                "$c_t$ is private consumption.\n",
                "\n",
                "$i_t$ is investment.\n",
                "\n",
                "$\\varepsilon_t$ is the random innovation to $z_t$\n",
                "\n",
                "We also have the following parameters:\n",
                "$\\alpha$ is the capital share in GDP.\n",
                "\n",
                "$\\beta$ is the subjective discount factor.\n",
                "\n",
                "$\\gamma$ is the curvature of the utility function.  In our example this is the constant elasticity of intertemporal substitition.  It is also the constant coefficient of relative risk aversion.\n",
                "\n",
                "$\\delta$ is the rate of capital depreciation.\n",
                "\n",
                "$\\rho$ is the autocorrelation of $z_t$.\n",
                "\n",
                "$\\sigma$ is the standard deviation of the $\\varepsilon_t$ shocks."
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "Suppose the household's Bellman equation is given by:\n",
                "$$ V(k_t, z_t) = \\max_{k_{t+1}} \\frac{c_t^{1-\\gamma}-1}{1-\\gamma} + \\beta V(k_{t+1}, z_{t+1}) $$\n",
                "\n",
                "$$ c_t = w_t + (1+r_t-\\delta)k_t - k_{t+1} $$\n",
                "\n",
                "The first-order condition is:\n",
                "$$ c_t^{-\\gamma} = \\beta V_k(k_{t+1}, z_{t+1}) $$\n",
                "The envelope condition is:\n",
                "$$ \\beta V_k(k_t, z_t) = c_t^{-\\gamma}(1+r_t-\\delta) $$\n",
                "Combining give the Euler equation:\n",
                "$$ c_t^{-\\gamma} = \\beta c_{t+1}^{-\\gamma}(1+r_{t+1}-\\delta) $$\n",
                "\n",
                "From the firm's problem we have the following:\n",
                "$$ y_t = k_t^\\alpha e^{(1-\\alpha)z_t} $$\n",
                "$$ r_t = \\alpha \\frac{y_t}{k_t} $$\n",
                "$$ w_t = (1-\\alpha)y_t $$\n",
                "\n",
                "Finally, we assume the following law of motion for $ z_t $:\n",
                "$$ z_t = \\rho z_{t-1} + \\varepsilon_t; \\varepsilon_t \\sim iid(0,\\sigma^2) $$"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "## Programming Definitions Functions"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "First we can define $y_t, r_t, w_t$ and $c_t$ as functions of the state $(k_t, z_t)$ using the budget constraint and the conditions from the firm's problem.  The parameters of the model, $\\alpha, \\beta, \\gamma, \\delta, \\rho, \\& \\sigma$, are included in the Python list, `mparams`.\n",
                "\n",
                "$$ y_t = k_t^\\alpha e^{(1-\\alpha)z_t} $$\n",
                "$$ r_t = \\alpha \\frac{y_t}{k_t} $$\n",
                "$$ w_t = (1-\\alpha)y_t $$\n",
                "$$ c_t = w_t + (1+r_t-\\delta)k_t - k_{t+1} $$\n",
                "$$ i_t = y_t - c_t $$\n",
                "$$ u_t = \\frac{c_t^{1-\\gamma}-1}{1-\\gamma} $$\n",
                "\n",
                "Note we need to consider the case where utility may be undefined.  In these cases we set it to a very large negative number."
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 1,
            "metadata": {},
            "outputs": [],
            "source": [
                "def modeldefs(Xm, Xn, Z, *mparams):\n",
                "    # unpack sets\n",
                "    k = Xm                            # k(t-1)\n",
                "    kp = Xn                           # k(t)\n",
                "    z = Z                             # z(t)\n",
                "    # find variables\n",
                "    y = k**alpha*np.exp((1-alpha)*z)  # y(t)\n",
                "    r = alpha*y/k                     # r(t)\n",
                "    w = (1-alpha)*y                   # w(t)\n",
                "    c = w + (1+r-delta)*k - kp        # c(t)\n",
                "    i = y - c                         # i(t)\n",
                "    if gamma == 1.0:                  # u(t)\n",
                "        u = np.log(c)\n",
                "    else:\n",
                "        u = (c**(1-gamma)-1)/(1-gamma)\n",
                "    if np.isnan(u) or np.isinf(u):\n",
                "        u = -1.0E+99\n",
                "    return y, r, w, c, i, u"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "## Writing the Main Program"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "Now we can write a program that will solve and simulate our model.\n",
                "First we need to call the Python packages and files we need."
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 3,
            "metadata": {},
            "outputs": [],
            "source": [
                "import numpy as np\n",
                "import scipy as sp\n",
                "import matplotlib.pyplot as plt\n",
                "from mpl_toolkits.mplot3d import Axes3D\n",
                "from rouwen import rouwen"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "Next we set our parameter values."
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 4,
            "metadata": {},
            "outputs": [],
            "source": [
                "alpha = .33\n",
                "beta = .95\n",
                "gamma = 2.5\n",
                "delta = .08\n",
                "rho = .9\n",
                "sigma = .02\n",
                "mparams = (alpha, beta, gamma, delta, rho, sigma)"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "### Finding the Steady State"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "With this mode it is straightforward to find an analytical solution for the steady state.\n",
                "\n",
                "For more complex models this may not be possible and we may need to use numerical techniques.  One way to find a numerical steady state is to use the using the linearization techniques implemented in the `LinApp` package.  This package can be downloaded at this [Github site](https://github.com/kerkphil/DSGE-Utilities). \n",
                "\n",
                "To find the steady state use the steady state version of the Euler equation to find $\\bar r$\n",
                "$$ \\bar c^{-\\gamma} = \\beta \\bar c^{-\\gamma}(1+\\bar r_-\\delta) $$\n",
                "$$ \\bar r = \\frac{1}{\\beta} + \\delta - 1 $$\n",
                "\n",
                "Next, use the first-order condition on capital from the firm's problem to find $\\bar k$.\n",
                "$$ \\bar r = \\alpha \\bar k^{\\alpha-1} e^{(1-\\alpha)0} $$\n",
                "$$ \\bar k = \\left( \\frac{\\alpha}{\\bar r} \\right)^{\\frac{1}{1-\\alpha}} $$"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 5,
            "metadata": {},
            "outputs": [
                {
                    "name": "stdout",
                    "output_type": "stream",
                    "text": [
                        "3.8980397970046203\n"
                    ]
                }
            ],
            "source": [
                "# find the steady state\n",
                "rbar = 1/beta + delta - 1\n",
                "kbar = (alpha/rbar)**(1/(1-alpha))\n",
                "print(kbar)"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "We can get the steady states for all the other variables using our definintions function"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 6,
            "metadata": {},
            "outputs": [
                {
                    "name": "stdout",
                    "output_type": "stream",
                    "text": [
                        "ybar:  1.5666762820497038\n",
                        "rbar:  0.13263157894736843\n",
                        "wbar:  1.0496731089733013\n",
                        "cbar:  1.2548330982893345\n",
                        "ibar:  0.31184318376036924\n",
                        "ubar:  0.19239214703970364\n"
                    ]
                }
            ],
            "source": [
                "zbar = 0.\n",
                "ybar, rbar, wbar, cbar, ibar, ubar = modeldefs(kbar, kbar, zbar, *mparams)\n",
                "print('ybar: ', ybar)\n",
                "print('rbar: ', rbar)\n",
                "print('wbar: ', wbar)\n",
                "print('cbar: ', cbar)\n",
                "print('ibar: ', ibar)\n",
                "print('ubar: ', ubar)"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "### Value Function Iteration"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "The next step is to find value function and policy function, $ k_{t+1} = \\Phi(k_t)$, by iteration on a discrete grid of points for $k$ and $z$.\n",
                "\n",
                "We begin by setting up grids using the steady state values as central points in the grid.  This means we need an odd number of points in the grid.  Having a value of $k=0$, can lead to undefined numbers, so we mke sure that spread of $k$ is such that all values are greater than zero (`keps` > 0 in the code).  For $k$ we set up a grid between $.01 \\bar k$ and $1.99 \\bar k$, by setting `keps` to .01.  "
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 7,
            "metadata": {},
            "outputs": [],
            "source": [
                "# set up grid for k\n",
                "keps = .01\n",
                "klow = keps*kbar\n",
                "khigh = (2-keps)*kbar\n",
                "knpts = 31\n",
                "kgrid = np.linspace(klow, khigh, num = knpts)"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "For $z$, our grid is between $-s\\sigma$ and $s\\sigma$, where we set $s$ as `spread`.\n",
                "\n",
                "In addition to the grid for $z$, it is also useful to approximate the AR(1) law of motion for $z$ with a Markov process.  The method for doing this was developed by [Tauchen (1986)](http://www.sciencedirect.com/science/article/pii/0165176586901680) and Rouwenhorst (1995).  We use Python code from Ben Tengelson based on MATLAB code from Lu Zhang and Karen Kopecky, which is avialable at [python4econ.blogspot.com](http://python4econ.blogspot.com/2013/10/rouwenhorst-1995-markov-transition.html).\n",
                "\n",
                "`rouwen.py` is an external python file containing the function, `rouwen(rho, mu, step, num)`.  We must import this by including **`from`** `rouwen` **`import`** `rouwen` at the begining of our program.\n",
                "\n",
                "This code generates both a grid for $z$ and the Markov transition probabilities of moving from one gridpoint to another from one period to the next.  Denoting the probability of moving from $z_i$ to $z_j$ as $\\pi_{ij}$, the matrix is organized with the $i$ corresponding to columns and the $j$ corresponding to rows, so that the columns all sum to one.\n",
                "\n",
                "We use these probabilities later to take the expected values in the Euler equation over $z_{t+1}$, given a value for $z_t$."
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 8,
            "metadata": {},
            "outputs": [],
            "source": [
                "# set up Markov approximation of AR(1) process using Rouwenhorst method\n",
                "spread = 2.  # number of standard deviations above and below 0\n",
                "znpts = 11\n",
                "zstep = 4.*spread*sigma/(znpts-1)\n",
                "# Markov transition probabilities, current z in cols, next z in rows\n",
                "Pimat, zgrid = rouwen(rho, 0., zstep, znpts)"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "Our value function, $V(k_t, z_t)$, is a two-dimensional array with rows corresponding to elements of the $k$ grid, and columns corresponding to the elements of $z$.  We store the value in the cells.\n",
                "\n",
                "Our policy function, $ k_{t+1} = \\Phi(k_t)$, is also a a two-dimensional array with the same rows and columns and we store an the optimal value of $k$ from the $k$ grid in the cells.\n",
                "\n",
                "We first initialize these arrays.  Note we need a place to hold our updated value function with our iterative procedure.  We call this `VFnew`."
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 9,
            "metadata": {},
            "outputs": [],
            "source": [
                "VF = np.zeros((knpts, znpts))\n",
                "VFnew = np.zeros((knpts, znpts))\n",
                "PF = np.zeros((knpts, znpts))"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "We need to set parameter values for the iterative prodcedure.\n",
                "\n",
                "`ccrit` tells us when the distance between `VF` and `VFnew` is small enough that we have converged to the true value function.  For this example we have set `crit = 1.0E-2` to ensure that the program executes quickly.  A more reasonable value would be something like `crit = 1.0E-10`, but this could take a long time to converge, particularly if the number of grid points is large.\n",
                "\n",
                "`maxit` is the maximum number of iterations we will allows.  If we have not found the true value fuction by this point, we will terminate the procedure.\n",
                "\n",
                "`damp` is the weight we put on the new value function relative to the old when moving to the next iteration.  Replacement of the old with the new corresponds to `damp = 1.`\n",
                "\n",
                "`dist` is the initial distance measure and needs to be larger than `ccrit`.\n",
                "\n",
                "`iters` is the initial count for the number of iterations peformed."
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 10,
            "metadata": {},
            "outputs": [],
            "source": [
                "ccrit = 1.0E-8\n",
                "maxit = 1000\n",
                "damp = 1.\n",
                "dist = 1.0E+99\n",
                "iters = 0"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "We next being a while loop where we find a new value for each point in the `VFnew` array.  We compare this to the original `VF` array and terminate if the two are sufficiently close or if we have exceeded the maximum number of iterations.\n",
                "\n",
                "To find the new values we loop over every possible value of $k_t$ and $z_t$ and calculate the maximimm value of $u(k_t, z_t) + \\beta E\\{V(k_{t+1}, z_{t+1}) \\} $, searching over every possible value of $k_{t+1}$.  We take the expected value over $z_{t+1}$ using the Markov transition probabilities we found earlier and placed in the array, `Pimat`.\n",
                "\n",
                "For our distance measure we will use either 1) the **mean** absolute deviation of `VF` from `VFnew` or 2) the **maximum** absolute deviation.  The latter is obviously a more restrictive measure of distance."
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 11,
            "metadata": {},
            "outputs": [
                {
                    "name": "stderr",
                    "output_type": "stream",
                    "text": [
                        "C:\\Users\\kerkp\\AppData\\Local\\Continuum\\anaconda3\\lib\\site-packages\\ipykernel_launcher.py:15: RuntimeWarning: invalid value encountered in double_scalars\n",
                        "  from ipykernel import kernelapp as app\n"
                    ]
                },
                {
                    "name": "stdout",
                    "output_type": "stream",
                    "text": [
                        "iteration:  1 distance:  32.84883119891331\n",
                        "iteration:  2 distance:  30.9001396693646\n",
                        "iteration:  3 distance:  29.095097972658124\n",
                        "iteration:  4 distance:  27.41939720949412\n",
                        "iteration:  5 distance:  25.860577817452466\n",
                        "iteration:  6 distance:  24.4077487136866\n",
                        "iteration:  7 distance:  23.051353572554746\n",
                        "iteration:  8 distance:  21.78297567374858\n",
                        "iteration:  9 distance:  20.59517443427498\n",
                        "iteration:  10 distance:  19.481348062024665\n",
                        "iteration:  11 distance:  18.435617819215793\n",
                        "iteration:  12 distance:  17.45273022130823\n",
                        "iteration:  13 distance:  16.52797416739054\n",
                        "iteration:  14 distance:  15.657110536816958\n",
                        "iteration:  15 distance:  14.836312165123388\n",
                        "iteration:  16 distance:  14.06210938194647\n",
                        "iteration:  17 distance:  13.33126632565046\n",
                        "iteration:  18 distance:  12.639968931939507\n",
                        "iteration:  19 distance:  11.981137386338673\n",
                        "iteration:  20 distance:  11.334417139345703\n",
                        "iteration:  21 distance:  10.654564760719893\n",
                        "iteration:  22 distance:  9.882747389827898\n",
                        "iteration:  23 distance:  8.962118760488522\n",
                        "iteration:  24 distance:  7.8481219522486185\n",
                        "iteration:  25 distance:  6.629886215763008\n",
                        "iteration:  26 distance:  5.433777185218332\n",
                        "iteration:  27 distance:  4.3117576976237615\n",
                        "iteration:  28 distance:  3.253067936048467\n",
                        "iteration:  29 distance:  2.338291319185089\n",
                        "iteration:  30 distance:  1.6257311599529771\n",
                        "iteration:  31 distance:  1.0998255771700656\n",
                        "iteration:  32 distance:  0.7236036680334337\n",
                        "iteration:  33 distance:  0.4599626892882611\n",
                        "iteration:  34 distance:  0.2778893205359054\n",
                        "iteration:  35 distance:  0.23515837495431668\n",
                        "iteration:  36 distance:  0.22247924294275379\n",
                        "iteration:  37 distance:  0.21056441646725688\n",
                        "iteration:  38 distance:  0.19935659449896548\n",
                        "iteration:  39 distance:  0.1888051707777123\n",
                        "iteration:  40 distance:  0.17886906064905617\n",
                        "iteration:  41 distance:  0.16949993213951586\n",
                        "iteration:  42 distance:  0.16065936696571725\n",
                        "iteration:  43 distance:  0.15231230791450656\n",
                        "iteration:  44 distance:  0.14442669194336588\n",
                        "iteration:  45 distance:  0.1369731248870707\n",
                        "iteration:  46 distance:  0.12992459553439062\n",
                        "iteration:  47 distance:  0.12325622505063905\n",
                        "iteration:  48 distance:  0.11694504746999712\n",
                        "iteration:  49 distance:  0.11096981725752642\n",
                        "iteration:  50 distance:  0.10531084037564398\n",
                        "iteration:  51 distance:  0.09994982573464739\n",
                        "iteration:  52 distance:  0.09486975431876488\n",
                        "iteration:  53 distance:  0.09005479554554086\n",
                        "iteration:  54 distance:  0.08549012135091516\n",
                        "iteration:  55 distance:  0.08116188426627957\n",
                        "iteration:  56 distance:  0.07705712269642362\n",
                        "iteration:  57 distance:  0.07316368587375166\n",
                        "iteration:  58 distance:  0.06947016967664155\n",
                        "iteration:  59 distance:  0.0659658591286103\n",
                        "iteration:  60 distance:  0.06264067673292573\n",
                        "iteration:  61 distance:  0.05948513591393123\n",
                        "iteration:  62 distance:  0.05649029893637447\n",
                        "iteration:  63 distance:  0.05364773876097928\n",
                        "iteration:  64 distance:  0.050949504368027476\n",
                        "iteration:  65 distance:  0.04838808914514203\n",
                        "iteration:  66 distance:  0.045956401989982965\n",
                        "iteration:  67 distance:  0.04364774082478107\n",
                        "iteration:  68 distance:  0.04145576826045527\n",
                        "iteration:  69 distance:  0.03937448918193098\n",
                        "iteration:  70 distance:  0.037398230055288195\n",
                        "iteration:  71 distance:  0.03552161978355173\n",
                        "iteration:  72 distance:  0.03373957195844923\n",
                        "iteration:  73 distance:  0.03204726837509497\n",
                        "iteration:  74 distance:  0.03044014369153558\n",
                        "iteration:  75 distance:  0.028913871130257007\n",
                        "iteration:  76 distance:  0.027464349129076027\n",
                        "iteration:  77 distance:  0.026087688861061054\n",
                        "iteration:  78 distance:  0.0247802025502466\n",
                        "iteration:  79 distance:  0.023538392519384388\n",
                        "iteration:  80 distance:  0.02235894091132451\n",
                        "iteration:  81 distance:  0.021238700032981228\n",
                        "iteration:  82 distance:  0.020174683274724714\n",
                        "iteration:  83 distance:  0.019164056563404586\n",
                        "iteration:  84 distance:  0.01820413031129675\n",
                        "iteration:  85 distance:  0.017292351825879848\n",
                        "iteration:  86 distance:  0.016426298149667673\n",
                        "iteration:  87 distance:  0.015603669301270884\n",
                        "iteration:  88 distance:  0.014822281891483158\n",
                        "iteration:  89 distance:  0.014080063090403883\n",
                        "iteration:  90 distance:  0.013375044923883195\n",
                        "iteration:  91 distance:  0.012705358878555728\n",
                        "iteration:  92 distance:  0.012069230797010277\n",
                        "iteration:  93 distance:  0.011464976045743924\n",
                        "iteration:  94 distance:  0.010890994939835252\n",
                        "iteration:  95 distance:  0.010345768409273148\n",
                        "iteration:  96 distance:  0.009827853893298766\n",
                        "iteration:  97 distance:  0.009335881449658245\n",
                        "iteration:  98 distance:  0.0088685500668646\n",
                        "iteration:  99 distance:  0.008424624167972183\n",
                        "iteration:  100 distance:  0.008002930295454291\n",
                        "iteration:  101 distance:  0.007602353967314457\n",
                        "iteration:  102 distance:  0.00722183669505938\n",
                        "iteration:  103 distance:  0.00686037315479382\n",
                        "iteration:  104 distance:  0.006517008503450938\n",
                        "iteration:  105 distance:  0.006190835832136088\n",
                        "iteration:  106 distance:  0.00588099374954254\n",
                        "iteration:  107 distance:  0.0055866640885895436\n",
                        "iteration:  108 distance:  0.005307069729596492\n",
                        "iteration:  109 distance:  0.005041472534294655\n",
                        "iteration:  110 distance:  0.004789171384501856\n",
                        "iteration:  111 distance:  0.004549500320401023\n",
                        "iteration:  112 distance:  0.004321826772752502\n",
                        "iteration:  113 distance:  0.004105549885139226\n",
                        "iteration:  114 distance:  0.0039000989203188396\n",
                        "iteration:  115 distance:  0.0037049317476558485\n",
                        "iteration:  116 distance:  0.003519533406432629\n",
                        "iteration:  117 distance:  0.003343414742047912\n",
                        "iteration:  118 distance:  0.0031761111105907958\n",
                        "iteration:  119 distance:  0.003017181148891268\n",
                        "iteration:  120 distance:  0.0028662056065016372\n",
                        "iteration:  121 distance:  0.00272278623627642\n",
                        "iteration:  122 distance:  0.0025865447410282627\n",
                        "iteration:  123 distance:  0.002457121773048243\n",
                        "iteration:  124 distance:  0.0023341759837620657\n",
                        "iteration:  125 distance:  0.002217383121525529\n",
                        "iteration:  126 distance:  0.0021064351742481335\n",
                        "iteration:  127 distance:  0.0020010395556866456\n",
                        "iteration:  128 distance:  0.0019009183319198542\n",
                        "iteration:  129 distance:  0.0018058074873366081\n",
                        "iteration:  130 distance:  0.0017154562268544282\n",
                        "iteration:  131 distance:  0.0016296263137576261\n",
                        "iteration:  132 distance:  0.0015480914401564405\n",
                        "iteration:  133 distance:  0.0014706366294916506\n",
                        "iteration:  134 distance:  0.0013970576688109304\n",
                        "iteration:  135 distance:  0.001327160569552177\n",
                        "iteration:  136 distance:  0.0012607610553558857\n",
                        "iteration:  137 distance:  0.0011976840755849594\n",
                        "iteration:  138 distance:  0.001137763343216136\n",
                        "iteration:  139 distance:  0.0010808408958808968\n",
                        "iteration:  140 distance:  0.0010267666790113594\n",
                        "iteration:  141 distance:  0.0009753981497127029\n",
                        "iteration:  142 distance:  0.0009265999006444758\n",
                        "iteration:  143 distance:  0.0008802433025891787\n",
                        "iteration:  144 distance:  0.0008362061652036346\n",
                        "iteration:  145 distance:  0.0007943724143970599\n",
                        "iteration:  146 distance:  0.0007546317861937268\n",
                        "iteration:  147 distance:  0.0007168795357941349\n",
                        "iteration:  148 distance:  0.0006810161611880972\n",
                        "iteration:  149 distance:  0.0006469471406731486\n",
                        "iteration:  150 distance:  0.0006145826835535217\n",
                        "iteration:  151 distance:  0.000583837493188355\n",
                        "iteration:  152 distance:  0.0005546305420409681\n",
                        "iteration:  153 distance:  0.0005268848579760288\n",
                        "iteration:  154 distance:  0.0005005273211793337\n",
                        "iteration:  155 distance:  0.00047548847134493144\n",
                        "iteration:  156 distance:  0.00045170232447588887\n",
                        "iteration:  157 distance:  0.00042910619885105916\n",
                        "iteration:  158 distance:  0.0004076405496675761\n",
                        "iteration:  159 distance:  0.00038724881216012363\n",
                        "iteration:  160 distance:  0.00036787725218800915\n",
                        "iteration:  161 distance:  0.000349474824865581\n",
                        "iteration:  162 distance:  0.0003319930396585846\n",
                        "iteration:  163 distance:  0.0003153858326427894\n",
                        "iteration:  164 distance:  0.0002996094450722353\n",
                        "iteration:  165 distance:  0.00028462230801551414\n",
                        "iteration:  166 distance:  0.00027038493262665497\n",
                        "iteration:  167 distance:  0.0002568598061571947\n",
                        "iteration:  168 distance:  0.0002440112930131022\n",
                        "iteration:  169 distance:  0.00023180554079260673\n",
                        "iteration:  170 distance:  0.00022021039099229256\n",
                        "iteration:  171 distance:  0.00020919529438856443\n",
                        "iteration:  172 distance:  0.00019873123031288742\n",
                        "iteration:  173 distance:  0.0001887906302187048\n",
                        "iteration:  174 distance:  0.00017934730510660302\n",
                        "iteration:  175 distance:  0.00017037637628902758\n",
                        "iteration:  176 distance:  0.00016185421000614042\n",
                        "iteration:  177 distance:  0.0001537583549691135\n",
                        "iteration:  178 distance:  0.00014606748329981656\n",
                        "iteration:  179 distance:  0.00013876133424162163\n",
                        "iteration:  180 distance:  0.00013182066084738153\n",
                        "iteration:  181 distance:  0.00012522717906193748\n",
                        "iteration:  182 distance:  0.00011896351973206265\n",
                        "iteration:  183 distance:  0.00011301318274803407\n",
                        "iteration:  184 distance:  0.00010736049353710087\n",
                        "iteration:  185 distance:  0.00010199056173831877\n",
                        "iteration:  186 distance:  9.688924205875082e-05\n",
                        "iteration:  187 distance:  9.204309686339229e-05\n",
                        "iteration:  188 distance:  8.743936084698589e-05\n",
                        "iteration:  189 distance:  8.306590738982322e-05\n",
                        "iteration:  190 distance:  7.891121655489997e-05\n",
                        "iteration:  191 distance:  7.496434474063562e-05\n",
                        "iteration:  192 distance:  7.121489589678731e-05\n",
                        "iteration:  193 distance:  6.765299404776215e-05\n",
                        "iteration:  194 distance:  6.426925735780742e-05\n",
                        "iteration:  195 distance:  6.10547732620148e-05\n",
                        "iteration:  196 distance:  5.800107520315123e-05\n",
                        "iteration:  197 distance:  5.5100120121664986e-05\n",
                        "iteration:  198 distance:  5.234426738809361e-05\n",
                        "iteration:  199 distance:  4.972625843180367e-05\n",
                        "iteration:  200 distance:  4.723919813898192e-05\n",
                        "iteration:  201 distance:  4.487653610141251e-05\n",
                        "iteration:  202 distance:  4.263204959897848e-05\n",
                        "iteration:  203 distance:  4.049982737797109e-05\n",
                        "iteration:  204 distance:  3.847425376335423e-05\n",
                        "iteration:  205 distance:  3.654999406421666e-05\n",
                        "iteration:  206 distance:  3.472198041976071e-05\n",
                        "iteration:  207 distance:  3.29853985903128e-05\n",
                        "iteration:  208 distance:  3.133567502544565e-05\n",
                        "iteration:  209 distance:  2.9768465104496045e-05\n",
                        "iteration:  210 distance:  2.82796414552422e-05\n",
                        "iteration:  211 distance:  2.6865283217603064e-05\n",
                        "iteration:  212 distance:  2.552166563418723e-05\n",
                        "iteration:  213 distance:  2.4245250301646593e-05\n",
                        "iteration:  214 distance:  2.3032675855461093e-05\n",
                        "iteration:  215 distance:  2.1880748981573106e-05\n",
                        "iteration:  216 distance:  2.0786436174091705e-05\n",
                        "iteration:  217 distance:  1.9746855656421758e-05\n",
                        "iteration:  218 distance:  1.8759269828194647e-05\n",
                        "iteration:  219 distance:  1.782107799641608e-05\n",
                        "iteration:  220 distance:  1.69298095471504e-05\n",
                        "iteration:  221 distance:  1.6083117515108825e-05\n",
                        "iteration:  222 distance:  1.527877227402996e-05\n",
                        "iteration:  223 distance:  1.451465571733479e-05\n",
                        "iteration:  224 distance:  1.3788755786947604e-05\n",
                        "iteration:  225 distance:  1.309916095948438e-05\n",
                        "iteration:  226 distance:  1.2444055343507898e-05\n",
                        "iteration:  227 distance:  1.1821713947313128e-05\n",
                        "iteration:  228 distance:  1.1230498024872304e-05\n",
                        "iteration:  229 distance:  1.0668850741524238e-05\n",
                        "iteration:  230 distance:  1.0135293244672994e-05\n",
                        "iteration:  231 distance:  9.628420571061724e-06\n",
                        "iteration:  232 distance:  9.14689805853186e-06\n",
                        "iteration:  233 distance:  8.689457800414857e-06\n",
                        "iteration:  234 distance:  8.254895341508472e-06\n",
                        "iteration:  235 distance:  7.842066423791039e-06\n",
                        "iteration:  236 distance:  7.449883987931116e-06\n",
                        "iteration:  237 distance:  7.077315537173945e-06\n",
                        "iteration:  238 distance:  6.72337994700456e-06\n",
                        "iteration:  239 distance:  6.387145361941293e-06\n",
                        "iteration:  240 distance:  6.067726502578807e-06\n",
                        "iteration:  241 distance:  5.764282278164501e-06\n",
                        "iteration:  242 distance:  5.4760137828679945e-06\n",
                        "iteration:  243 distance:  5.2021620007280944e-06\n",
                        "iteration:  244 distance:  4.9420058942928335e-06\n",
                        "iteration:  245 distance:  4.694860464837802e-06\n",
                        "iteration:  246 distance:  4.46007508259072e-06\n",
                        "iteration:  247 distance:  4.237031518528056e-06\n",
                        "iteration:  248 distance:  4.025142537500415e-06\n",
                        "iteration:  249 distance:  3.8238502568788135e-06\n",
                        "iteration:  250 distance:  3.6326247254692134e-06\n",
                        "iteration:  251 distance:  3.450962459794482e-06\n",
                        "iteration:  252 distance:  3.278385200644607e-06\n",
                        "iteration:  253 distance:  3.114438555940069e-06\n",
                        "iteration:  254 distance:  2.958690870968894e-06\n",
                        "iteration:  255 distance:  2.8107321696779763e-06\n",
                        "iteration:  256 distance:  2.670172861485298e-06\n",
                        "iteration:  257 distance:  2.536642881523221e-06\n",
                        "iteration:  258 distance:  2.409790667456946e-06\n",
                        "iteration:  259 distance:  2.289282306833229e-06\n",
                        "iteration:  260 distance:  2.1748004925825626e-06\n",
                        "iteration:  261 distance:  2.0660438408981463e-06\n",
                        "iteration:  262 distance:  1.9627260314791783e-06\n",
                        "iteration:  263 distance:  1.8645750401447003e-06\n",
                        "iteration:  264 distance:  1.7713325135559899e-06\n",
                        "iteration:  265 distance:  1.6827529307761324e-06\n",
                        "iteration:  266 distance:  1.5986031058901062e-06\n",
                        "iteration:  267 distance:  1.5186615200946108e-06\n",
                        "iteration:  268 distance:  1.4427176893150317e-06\n",
                        "iteration:  269 distance:  1.3705716952472358e-06\n",
                        "iteration:  270 distance:  1.3020336524505183e-06\n",
                        "iteration:  271 distance:  1.2369230546482868e-06\n",
                        "iteration:  272 distance:  1.1750685047218212e-06\n",
                        "iteration:  273 distance:  1.116307238646641e-06\n",
                        "iteration:  274 distance:  1.0604844788986156e-06\n",
                        "iteration:  275 distance:  1.0074533065562719e-06\n",
                        "iteration:  276 distance:  9.570741141828876e-07\n",
                        "iteration:  277 distance:  9.092142789768332e-07\n",
                        "iteration:  278 distance:  8.637478217110584e-07\n",
                        "iteration:  279 distance:  8.205549946183055e-07\n",
                        "iteration:  280 distance:  7.795221819151266e-07\n",
                        "iteration:  281 distance:  7.40541281629703e-07\n",
                        "iteration:  282 distance:  7.035097553398373e-07\n",
                        "iteration:  283 distance:  6.683300313170548e-07\n",
                        "iteration:  284 distance:  6.349095613700229e-07\n",
                        "iteration:  285 distance:  6.031603732026269e-07\n",
                        "iteration:  286 distance:  5.729988572511502e-07\n",
                        "iteration:  287 distance:  5.443456174702987e-07\n",
                        "iteration:  288 distance:  5.171252652758085e-07\n",
                        "iteration:  289 distance:  4.912660998002139e-07\n",
                        "iteration:  290 distance:  4.667000581548564e-07\n",
                        "iteration:  291 distance:  4.4336250937249133e-07\n",
                        "iteration:  292 distance:  4.211919701901934e-07\n",
                        "iteration:  293 distance:  4.0013011926021136e-07\n",
                        "iteration:  294 distance:  3.8012148451116445e-07\n",
                        "iteration:  295 distance:  3.6111341472633285e-07\n",
                        "iteration:  296 distance:  3.430558734862643e-07\n",
                        "iteration:  297 distance:  3.2590131837650915e-07\n",
                        "iteration:  298 distance:  3.096045944062098e-07\n",
                        "iteration:  299 distance:  2.941228132158358e-07\n",
                        "iteration:  300 distance:  2.794152180740639e-07\n",
                        "iteration:  301 distance:  2.6544306308551313e-07\n",
                        "iteration:  302 distance:  2.5216964161245414e-07\n",
                        "iteration:  303 distance:  2.39559945214296e-07\n",
                        "iteration:  304 distance:  2.2758079154527877e-07\n",
                        "iteration:  305 distance:  2.1620071777306293e-07\n",
                        "iteration:  306 distance:  2.0538965372907114e-07\n",
                        "iteration:  307 distance:  1.9511922033643714e-07\n",
                        "iteration:  308 distance:  1.8536238144406525e-07\n",
                        "iteration:  309 distance:  1.7609342961577568e-07\n",
                        "iteration:  310 distance:  1.6728797902487713e-07\n",
                        "iteration:  311 distance:  1.589228375564744e-07\n",
                        "iteration:  312 distance:  1.5097600680746837e-07\n",
                        "iteration:  313 distance:  1.4342656839971824e-07\n",
                        "iteration:  314 distance:  1.3625462713662273e-07\n",
                        "iteration:  315 distance:  1.2944131100312006e-07\n",
                        "iteration:  316 distance:  1.2296872142769644e-07\n",
                        "iteration:  317 distance:  1.1681976275212946e-07\n",
                        "iteration:  318 distance:  1.109783198671721e-07\n",
                        "iteration:  319 distance:  1.0542893846832158e-07\n",
                        "iteration:  320 distance:  1.0015706664034951e-07\n",
                        "iteration:  321 distance:  9.514883458905388e-08\n",
                        "iteration:  322 distance:  9.039104043040425e-08\n",
                        "iteration:  323 distance:  8.587110755797767e-08\n",
                        "iteration:  324 distance:  8.157724096236052e-08\n",
                        "iteration:  325 distance:  7.749807195978065e-08\n",
                        "iteration:  326 distance:  7.362287846035542e-08\n",
                        "iteration:  327 distance:  6.994144996497198e-08\n",
                        "iteration:  328 distance:  6.644414440870605e-08\n",
                        "iteration:  329 distance:  6.312168210342861e-08\n",
                        "iteration:  330 distance:  5.996537311148131e-08\n",
                        "iteration:  331 distance:  5.6966896977428405e-08\n",
                        "iteration:  332 distance:  5.4118359571475594e-08\n",
                        "iteration:  333 distance:  5.1412250456905895e-08\n",
                        "iteration:  334 distance:  4.884145710093435e-08\n",
                        "iteration:  335 distance:  4.6399236452998593e-08\n",
                        "iteration:  336 distance:  4.4079108363348496e-08\n",
                        "iteration:  337 distance:  4.187501190244802e-08\n",
                        "iteration:  338 distance:  3.9781134830718656e-08\n",
                        "iteration:  339 distance:  3.779193491482147e-08\n",
                        "iteration:  340 distance:  3.5902210981930693e-08\n",
                        "iteration:  341 distance:  3.410698923289601e-08\n",
                        "iteration:  342 distance:  3.240153745309726e-08\n",
                        "iteration:  343 distance:  3.07813721178718e-08\n",
                        "iteration:  344 distance:  2.9242201549095626e-08\n",
                        "iteration:  345 distance:  2.7779996969456988e-08\n",
                        "iteration:  346 distance:  2.6390914342755423e-08\n",
                        "iteration:  347 distance:  2.5071294373901765e-08\n",
                        "iteration:  348 distance:  2.381765540349079e-08\n",
                        "iteration:  349 distance:  2.2626714724083286e-08\n",
                        "iteration:  350 distance:  2.1495303315077763e-08\n",
                        "iteration:  351 distance:  2.0420479529548174e-08\n",
                        "iteration:  352 distance:  1.9399415407406195e-08\n",
                        "iteration:  353 distance:  1.8429382464546507e-08\n",
                        "iteration:  354 distance:  1.750785827425716e-08\n",
                        "iteration:  355 distance:  1.6632412780381856e-08\n",
                        "iteration:  356 distance:  1.580075092988409e-08\n",
                        "iteration:  357 distance:  1.501067714571036e-08\n",
                        "iteration:  358 distance:  1.4260109537644894e-08\n",
                        "iteration:  359 distance:  1.3547079902309633e-08\n",
                        "iteration:  360 distance:  1.2869676879745384e-08\n",
                        "iteration:  361 distance:  1.2226152534822177e-08\n",
                        "iteration:  362 distance:  1.1614815775828902e-08\n",
                        "iteration:  363 distance:  1.1034046565328026e-08\n",
                        "iteration:  364 distance:  1.0482324341865024e-08\n",
                        "iteration:  365 distance:  9.958185387404228e-09\n"
                    ]
                }
            ],
            "source": [
                "while (dist > ccrit) and (iters < maxit):\n",
                "    VFnew.fill(0.0)\n",
                "    iters = iters + 1\n",
                "    for i in range (0, knpts):\n",
                "        for j in range(0, znpts):\n",
                "            maxval = -1.0E+98\n",
                "            for m in range(0, knpts):\n",
                "                # get current period utility\n",
                "                yout, rat, wag, con, inv, u =  \\\n",
                "                    modeldefs(kgrid[i], kgrid[m], zgrid[j], *mparams)\n",
                "                # get expected value\n",
                "                val = 0.\n",
                "                for n in range (0, znpts):\n",
                "                    # sum over all possible value of z(t+1) with Markov probs\n",
                "                    val = val + u + beta*Pimat[n, j]*VF[m, n]\n",
                "                    # if this exceeds previous maximum do replacements\n",
                "                if val > maxval:\n",
                "                    maxval = val\n",
                "                    VFnew[i, j] = val\n",
                "                    PF[i, j] = kgrid[m]\n",
                "    # dist = np.mean(np.abs(VF - VFnew))\n",
                "    dist = np.amax(np.abs(VF - VFnew))\n",
                "    print('iteration: ', iters, 'distance: ', dist)\n",
                "    VF = damp*VFnew + (1-damp)*VF"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "We report the number of iterations it took to converge.\n",
                "\n",
                "We also check to see if the policy function has a steady state at $\\bar k$."
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 12,
            "metadata": {},
            "outputs": [
                {
                    "name": "stdout",
                    "output_type": "stream",
                    "text": [
                        "Converged after 365 iterations\n",
                        "Policy function at ( 15 , 5 ) should be 3.898039797004621 and is 3.898039797004621\n"
                    ]
                }
            ],
            "source": [
                "print('Converged after', iters, 'iterations')\n",
                "print('Policy function at (', int((knpts-1)/2), ',', int((znpts-1)/2), ') should be', \\\n",
                "    kgrid[int((knpts-1)/2)], 'and is', PF[int((knpts-1)/2), int((znpts-1)/2)])"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "### Fitting the Policy Function with a Polynomial"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "Our policy function is lumpy in the sense that the value of $k_t$, $z_t$ and $k_{t+1}$ are constrained to be the discrete values found in the grids.  This could be a problem when we simulate if the grids are not fine enough.  Adding more grid points slows down our value-function iteration.  One solution is to use a relativel coarse grid that will allow a quick solution for our value and policy functions, but to then fit the policy function with a continuous mutivariate polynomial in $k_t$ and $z_t$.  We do this below with a cubic polynomial.\n",
                "\n",
                "We first set up two-dimensional arrays for k and z, which give the value of these variables over the same space as the value function.  This is done using the meshgrid function in numpy.\n",
                "\n",
                "Next, we flatten the policy function from two dimensions to one and use this as the dependent variable in our regression.  For our independent variables we take appropriate powers of our k and z meshes, flatten them, and add them to the list of regressors.\n",
                "\n",
                "We then perform an OLS regression to generate fitted coefficients."
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 13,
            "metadata": {},
            "outputs": [],
            "source": [
                "# fit a polynomial\n",
                "# create meshgrid\n",
                "zmesh, kmesh = np.meshgrid(zgrid, kgrid)\n",
                "\n",
                "Y = PF.flatten()\n",
                "\n",
                "X = np.ones(knpts*znpts)\n",
                "\n",
                "temp = kmesh.flatten()\n",
                "X = np.vstack((X,temp))\n",
                "\n",
                "temp = kmesh**2\n",
                "temp = temp.flatten()\n",
                "X = np.vstack((X,temp))\n",
                "\n",
                "temp = kmesh**3\n",
                "temp = temp.flatten()\n",
                "X = np.vstack((X,temp))\n",
                "\n",
                "temp = zmesh.flatten()\n",
                "X = np.vstack((X,temp))\n",
                "\n",
                "temp = zmesh**2\n",
                "temp = temp.flatten()\n",
                "X = np.vstack((X,temp))\n",
                "\n",
                "temp = zmesh**3\n",
                "temp = temp.flatten()\n",
                "X = np.vstack((X,temp))\n",
                "\n",
                "temp = kmesh*zmesh\n",
                "temp = temp.flatten()\n",
                "X = np.vstack((X,temp))\n",
                "\n",
                "temp = kmesh**2*zmesh\n",
                "temp = temp.flatten()\n",
                "X = np.vstack((X,temp))\n",
                "\n",
                "temp = kmesh*zmesh**2\n",
                "temp = temp.flatten()\n",
                "X = np.vstack((X,temp))\n",
                "\n",
                "XtX = np.dot(X,np.transpose(X))\n",
                "XtY = np.dot(X,Y)\n",
                "coeffs = np.dot(np.linalg.inv(XtX),XtY)"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "With these coefficient estimates we can now produce a polynomial fitted version of the policy function.  We loop over all values of k and z in the grids and generate fitted values using the estimated coefficients.  We then compare these to the grid-based policy function to get a goodness of fit measure."
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 14,
            "metadata": {},
            "outputs": [
                {
                    "name": "stdout",
                    "output_type": "stream",
                    "text": [
                        "R-squared 0.9998048792600965\n"
                    ]
                }
            ],
            "source": [
                "PFpoly = np.zeros((knpts, znpts))\n",
                "for i in range(0, knpts):\n",
                "    for j in range(0, znpts):\n",
                "        PFpoly[i,j] = np.dot(np.stack((1, kgrid[i], kgrid[i]**2, kgrid[i]**3, \\\n",
                "        zgrid[j], zgrid[j]**2, zgrid[j]**3, \\\n",
                "        kgrid[i]*zgrid[j], kgrid[i]**2*zgrid[j], kgrid[i]*zgrid[j]**2)),coeffs)\n",
                "        \n",
                "# calcuate R-squared\n",
                "Rsq = 1 - np.sum((PF-PFpoly)**2)/np.sum(PF**2)\n",
                "print('R-squared', Rsq)"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "We can check how well our polynomial does at the steady state by putting in steady state values for k and z, and seeing how close it comes to generating kbar."
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 15,
            "metadata": {},
            "outputs": [
                {
                    "name": "stdout",
                    "output_type": "stream",
                    "text": [
                        "Value of Phi(kbar) from polynomial:  3.8947482173304167\n",
                        "kbar:  3.8980397970046203\n",
                        "Percent difference:  -0.00084441920699041\n",
                        "Absolute diffence:  -0.003291579674203682\n"
                    ]
                }
            ],
            "source": [
                "# check polynomial\n",
                "check = np.dot(np.stack((1, kbar, kbar**2, kbar**3, \\\n",
                "        zbar, zbar**2, zbar**3, kbar*zbar, \\\n",
                "        kbar**2*zbar, kbar*zbar**2)),coeffs)\n",
                "print('Value of Phi(kbar) from polynomial: ', check)\n",
                "print('kbar: ', kbar)\n",
                "print('Percent difference: ', check/kbar - 1.)\n",
                "diff = check - kbar\n",
                "print('Absolute diffence: ', diff)"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "Notice that the steady state value generated by the policy function is not the same as the steady state we found earlier.  This is because one we found first is the 'certainty equivalence' steady state.  That is the one to which the model would converge if there were never any stochastic shocks and the agents in the model knew this.  In reality our model has stochastic shocks and our agents are risk averse.  This leads to a higher level of savings because of a precautionary motive that does not exist in the certainty version of the model."
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "Lastly, we print surface plots of the grid-based value function, the grid-based policy function, and the polynomial policy function."
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 16,
            "metadata": {},
            "outputs": [
                {
                    "data": {
                        "image/png": "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\n",
                        "text/plain": [
                            "<Figure size 432x288 with 1 Axes>"
                        ]
                    },
                    "metadata": {
                        "needs_background": "light"
                    },
                    "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"
                },
                {
                    "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": [
                "# plot value function and transition function\n",
                "fig = plt.figure()\n",
                "ax = fig.add_subplot(111, projection='3d')\n",
                "ax.plot_surface(kmesh, zmesh, VF)\n",
                "ax.view_init(30, 150)\n",
                "plt.title('Value Function')\n",
                "plt.xlabel('k(t)')\n",
                "plt.ylabel('z(t)')\n",
                "plt.show()\n",
                "\n",
                "fig = plt.figure()\n",
                "ax = fig.add_subplot(111, projection='3d')\n",
                "ax.plot_surface(kmesh, zmesh, PF)\n",
                "ax.view_init(30, 150)\n",
                "plt.title('Transition Function Grid')\n",
                "plt.xlabel('k(t)')\n",
                "plt.ylabel('z(t)')\n",
                "plt.show()\n",
                "\n",
                "fig = plt.figure()\n",
                "ax = fig.add_subplot(111, projection='3d')\n",
                "ax.plot_surface(kmesh, zmesh, PFpoly)\n",
                "ax.view_init(30, 150)\n",
                "plt.title('Transition Function Polynomial')\n",
                "plt.xlabel('k(t)')\n",
                "plt.ylabel('z(t)')\n",
                "plt.show()"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "### Simulating the Model"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "Finally we simulate our model using the fitted polynomial\n"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 17,
            "metadata": {},
            "outputs": [],
            "source": [
                "# perform simulation\n",
                "T = 1000  # number of periods to simulate\n",
                "kstart = kbar # starting value for simulation\n",
                "# initialize variable histories\n",
                "epshist = np.random.randn(T)*sigma\n",
                "khist= np.zeros(T+1)\n",
                "zhist = np.zeros(T+1)\n",
                "yhist = np.zeros(T)\n",
                "rhist= np.zeros(T)\n",
                "whist = np.zeros(T)\n",
                "chist = np.zeros(T)\n",
                "ihist = np.zeros(T)\n",
                "uhist = np.zeros(T)\n",
                "khist[0] = kstart\n",
                "for t in range(0, T):\n",
                "    # perform simulation with polynomial fit\n",
                "    khist[t+1] = np.dot(np.stack((1, khist[t], khist[t]**2, khist[t]**3, \\\n",
                "        zhist[t], zhist[t]**2, zhist[t]**3, khist[t]*zhist[t], \\\n",
                "        khist[t]**2*zhist[t], khist[t]*zhist[t]**2)),coeffs)\n",
                "    zhist[t+1] = zhist[t]*rho + epshist[t]   \n",
                "    yhist[t], rhist[t], whist[t], chist[t], ihist[t], uhist[t] = \\\n",
                "        modeldefs(khist[t], khist[t+1], zhist[t], *mparams)   \n",
                "# remove final k & z\n",
                "khist = khist[0:T]\n",
                "zhist = zhist[0:T]"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "Lastly we can plot our simulated data."
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 18,
            "metadata": {
                "scrolled": true
            },
            "outputs": [
                {
                    "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": [
                "# plot data\n",
                "t = range(0, T)\n",
                "plt.plot(t, khist, label='k')\n",
                "plt.plot(t, zhist, label='z')\n",
                "plt.plot(t, yhist, label='y')\n",
                "plt.plot(t, rhist, label='r')\n",
                "plt.plot(t, whist, label='w')\n",
                "plt.plot(t, chist, label='c')\n",
                "plt.plot(t, ihist, label='i')\n",
                "plt.plot(t, uhist, label='u')\n",
                "plt.xlabel('time')\n",
                "plt.legend(loc=9, ncol=4, bbox_to_anchor=(0., 1.02, 1., .102))\n",
                "plt.show()"
            ]
        }
    ],
    "metadata": {
        "celltoolbar": "Raw Cell Format",
        "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": "3.7.3"
        }
    },
    "nbformat": 4,
    "nbformat_minor": 4
}