{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "b05d41ed-eeb0-4936-85ab-dbf12efc82d4",
   "metadata": {},
   "source": [
    "# Exponential smoothing in SQL"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1bbf544a-2be3-4579-ade3-404f4d850ef1",
   "metadata": {},
   "outputs": [],
   "source": [
    "from pandas import DataFrame\n",
    "import matplotlib.pyplot as plt\n",
    "from data201 import db_connection, df_query\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2dcedb2f-42ab-4b8a-bc71-8c9bdeb59793",
   "metadata": {},
   "outputs": [],
   "source": [
    "conn = db_connection(config_file='stock-prices.ini')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c08c9a81-7d93-4814-bbda-441a850331f0",
   "metadata": {},
   "outputs": [],
   "source": [
    "def compute_exponential_smoothings(conn, alpha, label):\n",
    "    \"\"\"\n",
    "    @conn the database connection\n",
    "    @alpha the alpha value for exponential smoothing\n",
    "    @label label for the exponentially smoothed values\n",
    "    \"\"\"\n",
    "    conn.cursor().execute('SET @smoothed = NULL')\n",
    "    \n",
    "    return df_query(conn,\n",
    "       f\"\"\" \n",
    "        SELECT days_ago, price,\n",
    "               @smoothed := ({alpha}*price\n",
    "                          + (1-{alpha})*(IF (@smoothed IS NULL, price, @smoothed)))\n",
    "                   AS '{label}'\n",
    "        FROM recent_prices\n",
    "        ORDER BY days_ago\n",
    "        \"\"\"\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8c1c0f1f-5020-48fc-bdb5-575d15df9791",
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.figure(figsize=(15, 10))\n",
    "\n",
    "first = True\n",
    "\n",
    "for a in [5, 2]:\n",
    "    alpha = a/10\n",
    "    graph_label = f'{alpha = }'\n",
    "    \n",
    "    df = compute_exponential_smoothings(conn, alpha, graph_label)\n",
    "    display(df)\n",
    "\n",
    "    rows = df.values.tolist()\n",
    "    xs   = [row[0] for row in rows]\n",
    "    ys   = [row[1] for row in rows]\n",
    "    exps = [row[2] for row in rows]\n",
    "    \n",
    "    # Initialize the graph.\n",
    "    if first:\n",
    "        plt.xticks(xs)\n",
    "        plt.plot(xs, ys, linewidth=5)\n",
    "        plt.title('Exponentially Smoothed Stock Prices')\n",
    "        plt.xlabel('Days ago')\n",
    "        plt.ylabel('Price')\n",
    "        first = False\n",
    "\n",
    "    # Plot the moving average line.\n",
    "    plt.plot(xs, exps, label=graph_label)\n",
    "    \n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b750ca51-c26e-43df-98b3-ec64a65eff16",
   "metadata": {},
   "outputs": [],
   "source": [
    "conn.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8e9ace8a-3452-456d-aa22-2c88df721822",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.12.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}