linear_additive_splitconformal.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "f30d1681",
   "metadata": {},
   "source": [
    "# Simulated data analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0cc4beb9",
   "metadata": {},
   "source": [
    "## Setup"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "1eca6bf7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "True coefficients: [-0.44329024  3.45930833 -0.17094764 -5.28926666 -3.07245074 -5.90810057\n",
      "  0.75680427]\n"
     ]
    }
   ],
   "source": [
    "# simulation setup\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "n = 100000\n",
    "p = 6\n",
    "np.random.seed(42)\n",
    "\n",
    "# fixed part\n",
    "x = np.random.rand(n, p)\n",
    "betas = np.random.normal(0, 5, size = p+1)\n",
    "print('True coefficients:', betas)\n",
    "true_f = betas[0] + np.dot(x, betas[1:])\n",
    "\n",
    "# stochastic part\n",
    "stn = 5\n",
    "noise_std = np.sqrt(np.var(true_f)/stn)\n",
    "noise = np.random.normal(0, noise_std, size = n)\n",
    "y = true_f + noise\n",
    "\n",
    "# save as pandas dataframe\n",
    "xy = pd.DataFrame(x, columns = [\"x\" + str(i+1) for i in range(p)])\n",
    "xy[\"y\"] = y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "149b5dbe",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>x1</th>\n",
       "      <th>x2</th>\n",
       "      <th>x3</th>\n",
       "      <th>x4</th>\n",
       "      <th>x5</th>\n",
       "      <th>x6</th>\n",
       "      <th>y</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.374540</td>\n",
       "      <td>0.950714</td>\n",
       "      <td>0.731994</td>\n",
       "      <td>0.598658</td>\n",
       "      <td>0.156019</td>\n",
       "      <td>0.155995</td>\n",
       "      <td>-5.469679</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.058084</td>\n",
       "      <td>0.866176</td>\n",
       "      <td>0.601115</td>\n",
       "      <td>0.708073</td>\n",
       "      <td>0.020584</td>\n",
       "      <td>0.969910</td>\n",
       "      <td>-3.749033</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.832443</td>\n",
       "      <td>0.212339</td>\n",
       "      <td>0.181825</td>\n",
       "      <td>0.183405</td>\n",
       "      <td>0.304242</td>\n",
       "      <td>0.524756</td>\n",
       "      <td>0.307718</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.431945</td>\n",
       "      <td>0.291229</td>\n",
       "      <td>0.611853</td>\n",
       "      <td>0.139494</td>\n",
       "      <td>0.292145</td>\n",
       "      <td>0.366362</td>\n",
       "      <td>-5.777635</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.456070</td>\n",
       "      <td>0.785176</td>\n",
       "      <td>0.199674</td>\n",
       "      <td>0.514234</td>\n",
       "      <td>0.592415</td>\n",
       "      <td>0.046450</td>\n",
       "      <td>-4.752492</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         x1        x2        x3        x4        x5        x6         y\n",
       "0  0.374540  0.950714  0.731994  0.598658  0.156019  0.155995 -5.469679\n",
       "1  0.058084  0.866176  0.601115  0.708073  0.020584  0.969910 -3.749033\n",
       "2  0.832443  0.212339  0.181825  0.183405  0.304242  0.524756  0.307718\n",
       "3  0.431945  0.291229  0.611853  0.139494  0.292145  0.366362 -5.777635\n",
       "4  0.456070  0.785176  0.199674  0.514234  0.592415  0.046450 -4.752492"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "xy.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "613428d3",
   "metadata": {},
   "outputs": [],
   "source": [
    "# spark setup\n",
    "import os\n",
    "import sys\n",
    "\n",
    "os.environ['PYSPARK_PYTHON'] = sys.executable\n",
    "os.environ['PYSPARK_DRIVER_PYTHON'] = sys.executable\n",
    "os.environ[\"PYARROW_IGNORE_TIMEZONE\"] = \"1\"\n",
    "\n",
    "# spark session\n",
    "from pyspark.sql import SparkSession\n",
    "spark = SparkSession \\\n",
    "    .builder \\\n",
    "    .master('local[4]') \\\n",
    "    .appName('This Spark session') \\\n",
    "    .config('spark.sql.repl.eagerEval.enabled', True) \\\n",
    "    .config('spark.sql.execution.arrow.pyspark.enabled', False) \\\n",
    "    .getOrCreate()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "51f6307c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "            <div>\n",
       "                <p><b>SparkSession - in-memory</b></p>\n",
       "                \n",
       "        <div>\n",
       "            <p><b>SparkContext</b></p>\n",
       "\n",
       "            <p><a href=\"http://DESKTOP-BDQVD8K:4040\">Spark UI</a></p>\n",
       "\n",
       "            <dl>\n",
       "              <dt>Version</dt>\n",
       "                <dd><code>v3.3.0</code></dd>\n",
       "              <dt>Master</dt>\n",
       "                <dd><code>local[4]</code></dd>\n",
       "              <dt>AppName</dt>\n",
       "                <dd><code>This Spark session</code></dd>\n",
       "            </dl>\n",
       "        </div>\n",
       "        \n",
       "            </div>\n",
       "        "
      ],
      "text/plain": [
       "<pyspark.sql.session.SparkSession at 0x25e89013880>"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "spark"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "0f0a915e",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-06T11:30:43.849057Z",
     "iopub.status.busy": "2022-10-06T11:30:43.848642Z",
     "iopub.status.idle": "2022-10-06T11:30:44.643075Z",
     "shell.execute_reply": "2022-10-06T11:30:44.642324Z",
     "shell.execute_reply.started": "2022-10-06T11:30:43.849019Z"
    }
   },
   "outputs": [],
   "source": [
    "# pyspark imports\n",
    "from pyspark.sql.functions import pandas_udf, col, lit, pow as spow\n",
    "from pyspark.ml.feature import VectorAssembler\n",
    "from pyspark.ml import Pipeline\n",
    "\n",
    "# other libraries\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "from scipy.linalg import solve_triangular"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b9ab3084",
   "metadata": {},
   "source": [
    "## Data preparation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "9a04ff6d",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-06T11:30:50.279266Z",
     "iopub.status.busy": "2022-10-06T11:30:50.279014Z",
     "iopub.status.idle": "2022-10-06T11:30:57.937070Z",
     "shell.execute_reply": "2022-10-06T11:30:57.936364Z",
     "shell.execute_reply.started": "2022-10-06T11:30:50.279238Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The dataset has 100000 rows and 7 columns.\n",
      "\n",
      "root\n",
      " |-- x1: double (nullable = true)\n",
      " |-- x2: double (nullable = true)\n",
      " |-- x3: double (nullable = true)\n",
      " |-- x4: double (nullable = true)\n",
      " |-- x5: double (nullable = true)\n",
      " |-- x6: double (nullable = true)\n",
      " |-- y: double (nullable = true)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# pyspark dataframe\n",
    "df = spark.createDataFrame(xy)\n",
    "\n",
    "n = df.count()\n",
    "print(f'The dataset has {n} rows and {len(df.columns)} columns.\\n')\n",
    "df.printSchema()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "155d0fb4",
   "metadata": {},
   "outputs": [],
   "source": [
    "# create feature preprocessing pipeline\n",
    "\n",
    "# combine features in a single vector\n",
    "features = df.columns[:-1]\n",
    "assembler = VectorAssembler(inputCols = features, outputCol = 'X')\n",
    "\n",
    "# put everything in a pipeline , to be used after data splitting\n",
    "preproc = Pipeline(stages = [assembler])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "1049b5dc",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-06T11:31:19.367884Z",
     "iopub.status.busy": "2022-10-06T11:31:19.367618Z",
     "iopub.status.idle": "2022-10-06T11:31:28.731765Z",
     "shell.execute_reply": "2022-10-06T11:31:28.731064Z",
     "shell.execute_reply.started": "2022-10-06T11:31:19.367847Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Each split has 49809 rows.\n"
     ]
    }
   ],
   "source": [
    "# obtain two equal splits\n",
    "s1, s2 = df.randomSplit([0.5, 0.5], seed=42)\n",
    "\n",
    "n1 = s1.count()\n",
    "n2 = s2.count()\n",
    "nn = min(n1, n2)\n",
    "\n",
    "s1 = s1.limit(nn)\n",
    "s2 = s2.limit(nn)\n",
    "\n",
    "print(f'Each split has {nn} rows.')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "d90a3333",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-06T11:31:28.733133Z",
     "iopub.status.busy": "2022-10-06T11:31:28.732884Z",
     "iopub.status.idle": "2022-10-06T11:32:08.207767Z",
     "shell.execute_reply": "2022-10-06T11:32:08.207086Z",
     "shell.execute_reply.started": "2022-10-06T11:31:28.733100Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "root\n",
      " |-- X: vector (nullable = true)\n",
      " |-- y: double (nullable = true)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# final preparation step\n",
    "s1 = preproc.fit(s1).transform(s1).select('X','y')\n",
    "s2 = preproc.fit(s2).transform(s2).select('X','y')\n",
    "\n",
    "s1.printSchema()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "869f1bdb",
   "metadata": {},
   "source": [
    "## Analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "85ab2d68",
   "metadata": {},
   "source": [
    "### Linear model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "88e7d560",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-06T11:32:08.209252Z",
     "iopub.status.busy": "2022-10-06T11:32:08.208976Z",
     "iopub.status.idle": "2022-10-06T11:32:08.264561Z",
     "shell.execute_reply": "2022-10-06T11:32:08.264017Z",
     "shell.execute_reply.started": "2022-10-06T11:32:08.209199Z"
    }
   },
   "outputs": [],
   "source": [
    "# functions for liner model estimation and prediction\n",
    "\n",
    "def map_lm(row):\n",
    "    \"\"\"Calculate x*t(x) and x*y for each row\"\"\"\n",
    "    \n",
    "    x = [1]\n",
    "    x.extend(row['X'].toArray())\n",
    "    x = np.array(x)\n",
    "    y = row[\"y\"]\n",
    "    out = [np.outer(x, x), x*y]    \n",
    "    \n",
    "    return out\n",
    "\n",
    "def get_prediction_lm(row, beta):\n",
    "    \"\"\"Calculate predictions for a row from estimated linear model\"\"\"\n",
    "    \n",
    "    x = row[\"X\"].toArray()\n",
    "    ypred = beta[1] + np.dot(x, beta[1:])\n",
    "    \n",
    "    return (row + (float(ypred),))\n",
    "\n",
    "def cholesky_normal_equations(xtx, xty):\n",
    "    \"\"\"Calculate the solution to the normal equation via Cholesky scomposition\"\"\"\n",
    "    \n",
    "    if not np.all(np.linalg.eigvals(xtx) > 0):\n",
    "        print('Adjusting matrix...')\n",
    "        #adj = np.abs(np.min(np.linalg.eigvals(xtx)))+0.00001\n",
    "        xtx = xtx + np.diag(np.repeat(0.001, xtx.shape[0]))\n",
    "    \n",
    "    L = np.linalg.cholesky(xtx) \n",
    "    z = solve_triangular(L, xty, lower = True) \n",
    "    beta_hat = solve_triangular(L, z, lower = True, trans = \"T\")\n",
    "    \n",
    "    return beta_hat    \n",
    "\n",
    "def get_beta_sd(xtx, sigma2):\n",
    "    '''\n",
    "    calculates standard deviation of linear model coefficients \n",
    "    using Cholesky decomposition\n",
    "    '''\n",
    "    \n",
    "    L = np.linalg.cholesky(xtx) \n",
    "    z1 = solve_triangular(L, np.eye(xtx.shape[0]), lower = True)\n",
    "    z2 = solve_triangular(L.T, np.eye(xtx.shape[0]), lower = False)\n",
    "    xtx_inv = np.dot(z2, z1)\n",
    "    sd_beta = np.sqrt(np.diag(sigma2*xtx_inv))\n",
    "    \n",
    "    return sd_beta"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "6d82c7e7",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-06T11:32:08.265563Z",
     "iopub.status.busy": "2022-10-06T11:32:08.265394Z",
     "iopub.status.idle": "2022-10-06T11:32:55.692806Z",
     "shell.execute_reply": "2022-10-06T11:32:55.692164Z",
     "shell.execute_reply.started": "2022-10-06T11:32:08.265542Z"
    }
   },
   "outputs": [],
   "source": [
    "# calculate t(X)X and t(X)y for the two splits\n",
    "xtx_lm_s1, xty_lm_s1 = s1.rdd.map(lambda row: map_lm(row)).reduce(lambda a,b: (a[0]+b[0], a[1]+b[1]))\n",
    "xtx_lm_s2, xty_lm_s2 = s2.rdd.map(lambda row: map_lm(row)).reduce(lambda a,b: (a[0]+b[0], a[1]+b[1]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "08adae44",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-06T11:32:55.693898Z",
     "iopub.status.busy": "2022-10-06T11:32:55.693728Z",
     "iopub.status.idle": "2022-10-06T11:32:55.744019Z",
     "shell.execute_reply": "2022-10-06T11:32:55.743333Z",
     "shell.execute_reply.started": "2022-10-06T11:32:55.693876Z"
    }
   },
   "outputs": [],
   "source": [
    "# solve normal equations\n",
    "beta_hat_lm_s1 = cholesky_normal_equations(xtx_lm_s1, xty_lm_s1)\n",
    "beta_hat_lm_s2 = cholesky_normal_equations(xtx_lm_s2, xty_lm_s2)\n",
    "\n",
    "#beta_hat_lm_s1, beta_hat_lm_s2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "dede1b63",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-06T11:32:55.803561Z",
     "iopub.status.busy": "2022-10-06T11:32:55.803374Z",
     "iopub.status.idle": "2022-10-06T11:32:56.564930Z",
     "shell.execute_reply": "2022-10-06T11:32:56.564252Z",
     "shell.execute_reply.started": "2022-10-06T11:32:55.803540Z"
    }
   },
   "outputs": [],
   "source": [
    "# calculate predictions from linear model\n",
    "o1 = s1.rdd.map(lambda row:get_prediction_lm(row, beta_hat_lm_s2)).toDF(['X', 'y', 'yhat'])\n",
    "o2 = s2.rdd.map(lambda row:get_prediction_lm(row, beta_hat_lm_s1)).toDF(['X', 'y', 'yhat'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "0d52a7ad",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-06T11:32:56.567322Z",
     "iopub.status.busy": "2022-10-06T11:32:56.567148Z",
     "iopub.status.idle": "2022-10-06T11:32:57.332204Z",
     "shell.execute_reply": "2022-10-06T11:32:57.331617Z",
     "shell.execute_reply.started": "2022-10-06T11:32:56.567301Z"
    }
   },
   "outputs": [],
   "source": [
    "# functions for conformal prediction\n",
    "\n",
    "@pandas_udf(\"double\")\n",
    "def get_abs_error(yreal: pd.Series, ypred: pd.Series) -> pd.Series:\n",
    "    \"\"\"calculate absolute prediction error\"\"\"\n",
    "    return np.abs(yreal-ypred)\n",
    "\n",
    "@pandas_udf(\"double\")\n",
    "def find_quantile(err: pd.Series) -> pd.Series:\n",
    "    \"\"\"given ordered absolute errors, calculate ROO empirical quantile\"\"\"\n",
    "    \n",
    "    level = 0.95\n",
    "    m = int(np.ceil(len(err)*level))\n",
    "    out = [err.drop([i]).values[m-1] for i in range(len(err))]\n",
    "    \n",
    "    return pd.Series(out)\n",
    "\n",
    "def ROOsplitconformal(df):\n",
    "    \"\"\"\n",
    "    given a Spark dataframe with columns y and yhat,\n",
    "    calculate and sort absolute errors, \n",
    "    find ROO empirical quantiles,\n",
    "    calculate lower and upper bounds for ROO split conformal prediction intervals    \n",
    "    \"\"\"\n",
    "    \n",
    "    out = df \\\n",
    "            .withColumn(\"abs_error\", get_abs_error(col(\"y\"), col(\"yhat\"))).sort(col(\"abs_error\")) \\\n",
    "            .withColumn(\"ROO_eq\", find_quantile(col(\"abs_error\"))) \\\n",
    "            .withColumn(\"lower\", col(\"yhat\")-col(\"ROO_eq\")) \\\n",
    "            .withColumn(\"upper\", col(\"yhat\")+col(\"ROO_eq\"))\n",
    "    \n",
    "    return out"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "d8f78886",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-06T11:32:57.333609Z",
     "iopub.status.busy": "2022-10-06T11:32:57.333429Z",
     "iopub.status.idle": "2022-10-06T11:32:57.598611Z",
     "shell.execute_reply": "2022-10-06T11:32:57.597915Z",
     "shell.execute_reply.started": "2022-10-06T11:32:57.333586Z"
    }
   },
   "outputs": [],
   "source": [
    "# ROO split conformal predictive intervals\n",
    "o1 = o1.transform(ROOsplitconformal)\n",
    "o2 = o2.transform(ROOsplitconformal)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "b0fa4c65",
   "metadata": {},
   "outputs": [],
   "source": [
    "# calculate betas standard error\n",
    "\n",
    "# get s2 for each split\n",
    "sigma2_hat1 = o1.withColumn('rsse', spow((col('yhat')-col('y')), 2)).select('rsse').groupBy().sum().collect()[0][0]/(nn-1)\n",
    "sigma2_hat2 = o2.withColumn('rsse', spow((col('yhat')-col('y')), 2)).select('rsse').groupBy().sum().collect()[0][0]/(nn-1)\n",
    "\n",
    "# get betas sd for each split\n",
    "beta_hat_lm_s1_sd = get_beta_sd(xtx_lm_s1, sigma2_hat1)\n",
    "beta_hat_lm_s2_sd = get_beta_sd(xtx_lm_s2, sigma2_hat2)\n",
    "\n",
    "#beta_hat_lm_s1_sd, beta_hat_lm_s2_sd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "05f8a903",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-06T11:32:57.333609Z",
     "iopub.status.busy": "2022-10-06T11:32:57.333429Z",
     "iopub.status.idle": "2022-10-06T11:32:57.598611Z",
     "shell.execute_reply": "2022-10-06T11:32:57.597915Z",
     "shell.execute_reply.started": "2022-10-06T11:32:57.333586Z"
    }
   },
   "outputs": [],
   "source": [
    "# merge the splits\n",
    "res_lm = o1.union(o2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "e9268612",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-06T11:32:57.599718Z",
     "iopub.status.busy": "2022-10-06T11:32:57.599536Z",
     "iopub.status.idle": "2022-10-06T11:32:57.862842Z",
     "shell.execute_reply": "2022-10-06T11:32:57.862229Z",
     "shell.execute_reply.started": "2022-10-06T11:32:57.599696Z"
    }
   },
   "outputs": [],
   "source": [
    "# functions for results evaluation\n",
    "\n",
    "@pandas_udf(\"double\")\n",
    "def rmse(yreal: pd.Series, ypred: pd.Series) -> float:\n",
    "    return np.sqrt(np.mean((yreal-ypred)**2))\n",
    "\n",
    "@pandas_udf(\"double\")\n",
    "def coverage(yreal: pd.Series, lower: pd.Series, upper: pd.Series) -> float:\n",
    "    return np.mean(((yreal>lower) & (yreal<upper))*1.0)\n",
    "\n",
    "@pandas_udf(\"double\")\n",
    "def avg_length(lower: pd.Series, upper: pd.Series) -> float:\n",
    "    return np.mean(upper - lower)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "3376b185",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-06T11:32:57.864228Z",
     "iopub.status.busy": "2022-10-06T11:32:57.863975Z",
     "iopub.status.idle": "2022-10-06T11:32:58.608715Z",
     "shell.execute_reply": "2022-10-06T11:32:58.601417Z",
     "shell.execute_reply.started": "2022-10-06T11:32:57.864193Z"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# evaluate results\n",
    "metrics_lm = res_lm.select(\n",
    "    lit('linear').alias('model'),\n",
    "    rmse(col('y'), col('yhat')).alias('rmse'),\n",
    "    coverage(col('y'), col('lower'), col('upper')).alias('coverage'), \n",
    "    avg_length(col('lower'), col('upper')).alias('avg_length'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "0450a0e0",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-06T11:32:58.609867Z",
     "iopub.status.busy": "2022-10-06T11:32:58.609684Z",
     "iopub.status.idle": "2022-10-06T11:32:58.666819Z",
     "shell.execute_reply": "2022-10-06T11:32:58.666130Z",
     "shell.execute_reply.started": "2022-10-06T11:32:58.609845Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "+------+-----------------+------------------+-----------------+\n",
      "| model|             rmse|          coverage|       avg_length|\n",
      "+------+-----------------+------------------+-----------------+\n",
      "|linear|4.088984186384552|0.9500090345118352|8.869898319485893|\n",
      "+------+-----------------+------------------+-----------------+\n",
      "\n"
     ]
    }
   ],
   "source": [
    "metrics_lm.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "930e8e52",
   "metadata": {},
   "source": [
    "### Additive model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "071ff57c",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-06T11:32:58.668215Z",
     "iopub.status.busy": "2022-10-06T11:32:58.667970Z",
     "iopub.status.idle": "2022-10-06T11:32:58.730290Z",
     "shell.execute_reply": "2022-10-06T11:32:58.729743Z",
     "shell.execute_reply.started": "2022-10-06T11:32:58.668183Z"
    }
   },
   "outputs": [],
   "source": [
    "# functions for additive model estimation and prediction\n",
    "\n",
    "def ncs_elem(x, xk):\n",
    "    \"\"\"given a feature and knots, calulates basis functtion\"\"\"\n",
    "    \n",
    "    out = (((xk-1/2)**2-1/12)*((x-1/2)**2)-1/12)/4 - ((np.abs(x-xk)-1/2)**4-1/2*(np.abs(x-xk)-1/2)**2 + 7/240)/24\n",
    "    \n",
    "    return out\n",
    "\n",
    "def splS(knots):\n",
    "    \"\"\"define S penalty matrix for a single term\"\"\"\n",
    "    \n",
    "    q = len(knots)+2\n",
    "    S = np.zeros([q,q])\n",
    "    k = np.array(knots)\n",
    "    S[2:, 2:] = ncs_elem(k[:,None], k)\n",
    "    \n",
    "    return S\n",
    "    \n",
    "def get_bigS(knots, lam = 0.01):\n",
    "    \"\"\"Obtain global penalty matrix\"\"\"\n",
    "    \n",
    "    Slist = []\n",
    "    for i in range(p):\n",
    "        q = len(knots)+1\n",
    "        S = np.zeros([p*q, p*q])\n",
    "        S2 = np.zeros([len(S)+1, len(S)+1])\n",
    "        S[i*q:(i+1)*q, i*q:(i+1)*q] = splS(knots)[1:,1:]\n",
    "        S2[1:,1:] = S\n",
    "        Slist.append(S2)\n",
    "    bigS = sum(lam*elem for elem in Slist)\n",
    "    \n",
    "    return bigS\n",
    "\n",
    "def map_gam(row, knots):\n",
    "    \"\"\"for each row, compute natural cubic splines basis expansion for x then calculate x*t(x) and x*y\"\"\"\n",
    "    \n",
    "    nx = [1]\n",
    "    x = row[\"X\"].toArray()\n",
    "    y = row[\"y\"]\n",
    "    for j, xij in enumerate(x):\n",
    "        nx.append(xij)\n",
    "        nx.extend([ncs_elem(xij, knot) for knot in knots])\n",
    "    x = np.array(nx)\n",
    "    out = [np.outer(x, x), x*y]\n",
    "\n",
    "    return out\n",
    "\n",
    "def get_prediction_gam(row, beta, knots):\n",
    "    \"\"\"Calculate predictions for estimated additive model\"\"\"\n",
    "    \n",
    "    nx = []\n",
    "    x = row[\"X\"].toArray()\n",
    "    for j, xij in enumerate(x):\n",
    "        nx.append(xij)\n",
    "        nx.extend([ncs_elem(xij, knot) for knot in knots])\n",
    "    x = np.array(nx)\n",
    "    ypred = beta[1] + np.dot(x, beta[1:])\n",
    "    \n",
    "    return (row + (float(ypred),))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "332f25e5",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-06T11:32:58.731344Z",
     "iopub.status.busy": "2022-10-06T11:32:58.731176Z",
     "iopub.status.idle": "2022-10-06T11:32:58.783326Z",
     "shell.execute_reply": "2022-10-06T11:32:58.782607Z",
     "shell.execute_reply.started": "2022-10-06T11:32:58.731323Z"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "knots1 = np.linspace(0.05, 0.95, 10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "724089d4",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-06T11:32:58.784689Z",
     "iopub.status.busy": "2022-10-06T11:32:58.784436Z",
     "iopub.status.idle": "2022-10-06T11:41:01.244720Z",
     "shell.execute_reply": "2022-10-06T11:41:01.244082Z",
     "shell.execute_reply.started": "2022-10-06T11:32:58.784654Z"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# calculate t(X)X and t(X)y for the two splits\n",
    "xtx_gam1_s1, xty_gam1_s1 = s1.rdd.map(lambda row: map_gam(row, knots1)).reduce(lambda a,b: (a[0]+b[0], a[1]+b[1]))\n",
    "xtx_gam1_s2, xty_gam1_s2 = s2.rdd.map(lambda row: map_gam(row, knots1)).reduce(lambda a,b: (a[0]+b[0], a[1]+b[1]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "9d6d3e24",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-06T11:41:01.245682Z",
     "iopub.status.busy": "2022-10-06T11:41:01.245511Z",
     "iopub.status.idle": "2022-10-06T11:41:01.300338Z",
     "shell.execute_reply": "2022-10-06T11:41:01.299822Z",
     "shell.execute_reply.started": "2022-10-06T11:41:01.245661Z"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# solve normal equations using LU decomposition since penalized matrix is not symmetric\n",
    "S = get_bigS(knots = knots1, lam = 4)\n",
    "beta_hat_gam1_s1 = np.linalg.solve(xtx_gam1_s1 + S, xty_gam1_s1)\n",
    "beta_hat_gam1_s2 = np.linalg.solve(xtx_gam1_s2 + S, xty_gam1_s2)\n",
    "\n",
    "#beta_hat_gam1_s1, beta_hat_gam1_s2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "85cd9db8",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-06T11:41:01.352351Z",
     "iopub.status.busy": "2022-10-06T11:41:01.352170Z",
     "iopub.status.idle": "2022-10-06T11:41:02.111672Z",
     "shell.execute_reply": "2022-10-06T11:41:02.110934Z",
     "shell.execute_reply.started": "2022-10-06T11:41:01.352329Z"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# calculate predictions from additive model\n",
    "o1 = s1.rdd.map(lambda row:get_prediction_gam(row, beta_hat_gam1_s2, knots1)).toDF(['X', 'y', 'yhat'])\n",
    "o2 = s2.rdd.map(lambda row:get_prediction_gam(row, beta_hat_gam1_s1, knots1)).toDF(['X', 'y', 'yhat'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "6a0e1cda",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-06T11:41:02.112741Z",
     "iopub.status.busy": "2022-10-06T11:41:02.112563Z",
     "iopub.status.idle": "2022-10-06T11:41:02.366960Z",
     "shell.execute_reply": "2022-10-06T11:41:02.366348Z",
     "shell.execute_reply.started": "2022-10-06T11:41:02.112716Z"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# ROO split conformal predictive intervals\n",
    "o1 = o1.transform(ROOsplitconformal)\n",
    "o2 = o2.transform(ROOsplitconformal)\n",
    "\n",
    "# merge the splits\n",
    "res_gam1 = o1.union(o2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "42967d41",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-06T11:41:02.367983Z",
     "iopub.status.busy": "2022-10-06T11:41:02.367775Z",
     "iopub.status.idle": "2022-10-06T11:41:02.624856Z",
     "shell.execute_reply": "2022-10-06T11:41:02.624167Z",
     "shell.execute_reply.started": "2022-10-06T11:41:02.367960Z"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# evaluate results\n",
    "metrics_gam1 = res_gam1.select(\n",
    "    lit('additive').alias('model'),\n",
    "    rmse(col('y'), col('yhat')).alias('rmse'),\n",
    "    coverage(col('y'), col('lower'), col('upper')).alias('coverage'), \n",
    "    avg_length(col('lower'), col('upper')).alias('avg_length'))\n",
    "\n",
    "metrics = metrics_lm.union(metrics_gam1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "56cb4317",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-06T11:41:02.625870Z",
     "iopub.status.busy": "2022-10-06T11:41:02.625703Z",
     "iopub.status.idle": "2022-10-06T11:57:32.677760Z",
     "shell.execute_reply": "2022-10-06T11:57:32.677174Z",
     "shell.execute_reply.started": "2022-10-06T11:41:02.625848Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "+--------+-----------------+------------------+-----------------+\n",
      "|   model|             rmse|          coverage|       avg_length|\n",
      "+--------+-----------------+------------------+-----------------+\n",
      "|  linear|4.088984186384552|0.9500090345118352|8.869898319485893|\n",
      "|additive|4.094623194083175|0.9500090345118352|8.881213647532158|\n",
      "+--------+-----------------+------------------+-----------------+\n",
      "\n"
     ]
    }
   ],
   "source": [
    "metrics.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cfdc4b34",
   "metadata": {},
   "source": [
    "## Visualization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "9233c991",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "# helper functions\n",
    "knots = np.linspace(0.05, 0.95, 10)\n",
    "num = [\"x\" + str(i+1) for i in range(p)]\n",
    "\n",
    "def plot_lm(beta_hat_lm, beta_hat_lm_sd):\n",
    "\n",
    "    fig, ax = plt.subplots(figsize=(1.618*5, 5))\n",
    "    fig.suptitle('Linear model - Estimated coefficients', size=18)\n",
    "    sns.barplot(ax=ax, y=num, x=beta_hat_lm[1:], orient='h', color='tab:blue')\n",
    "    y_coords = [p.get_y() + 0.5 * p.get_height() for p in ax.patches]\n",
    "    x_coords = [p.get_width() for p in ax.patches]\n",
    "    ax.errorbar(x=x_coords, y=y_coords, xerr=beta_hat_lm_sd[1:], fmt=\"none\", c=\"k\")\n",
    "    ax.axvline(0, 0, 1, linestyle='dashed', linewidth=0.8, color='darkgrey')\n",
    "    fig.tight_layout()\n",
    "\n",
    "def ncs_elem(x, xk):\n",
    "    \"\"\"given a feature and knots, calulates basis functtion\"\"\"\n",
    "    out = (((xk-1/2)**2-1/12)*((x-1/2)**2)-1/12)/4 - ((np.abs(x-xk)-1/2)**4-1/2*(np.abs(x-xk)-1/2)**2 + 7/240)/24\n",
    "    return out\n",
    "    \n",
    "def plot_gam(gam_coefs, sharey=False):\n",
    "    \n",
    "    coefs = np.split(np.delete(gam_coefs, [0]), 6)\n",
    "\n",
    "    x = np.linspace(0, 1, 100)\n",
    "    nx = []\n",
    "    nx.append(x)\n",
    "    nx.extend([ncs_elem(x, knot) for knot in knots])\n",
    "    nx = np.array(nx).T\n",
    "\n",
    "    fig, axs = plt.subplots(2, 3, figsize = [1.618*6, 6], sharex=True, sharey=sharey)\n",
    "    fig.suptitle('Additive model - Estimated effects', size=18)\n",
    "    for i, ax in enumerate(axs.ravel()):\n",
    "        y_gam = np.dot(nx, coefs[i])\n",
    "        plotdata = pd.DataFrame({'x':x, 'y':y_gam})\n",
    "        sns.lineplot(ax=ax, x='x', y='y', data=plotdata)\n",
    "        ax.set_title(num[i], size=12)\n",
    "    \n",
    "    plt.show()\n",
    "       \n",
    "def plot_gam_lm(gam_coefs, lin_coefs, lin_coefs_sd, sharey=False, height=6, save=False):\n",
    "    \n",
    "    coefs = np.split(np.delete(gam_coefs, [0]), 6)\n",
    "\n",
    "    x = np.linspace(0, 1, 100)\n",
    "    nx = []\n",
    "    nx.append(x)\n",
    "    nx.extend([ncs_elem(x, knot) for knot in knots])\n",
    "    nx = np.array(nx).T\n",
    "\n",
    "    fig, axs = plt.subplots(2, 3, figsize = (1.618*height, height), sharex=True, sharey=sharey)\n",
    "    fig.suptitle('Additive model - Estimated effects', size=18)\n",
    "    for i, ax in enumerate(axs.ravel()):\n",
    "        y_gam = np.dot(nx, coefs[i])\n",
    "        y_lin = np.dot(x, lin_coefs[i+1])\n",
    "        y1 = np.dot(x, lin_coefs[i+1] - 1.96* lin_coefs_sd[i+1])\n",
    "        y2 = np.dot(x, lin_coefs[i+1] + 1.96* lin_coefs_sd[i+1])\n",
    "        plotdata = pd.DataFrame({'x':x, 'additive':y_gam, 'linear':y_lin})\n",
    "        sns.lineplot(ax=ax, x='x', y='y', hue='model', data=pd.melt(plotdata, id_vars='x', var_name='model', value_name='y'))\n",
    "        ax.fill_between(x=plotdata.x, y1=y1, y2=y2, color='orange', alpha=0.1)\n",
    "        ax.axhline(0, 0, 1, linestyle='dashed', linewidth=0.8, color='darkgrey')\n",
    "        ax.set_title(num[i], size=12)\n",
    "        ax.set(xlabel='', ylabel='')\n",
    "        ax.get_legend().remove()\n",
    "    \n",
    "    handles, labels = axs[1,1].get_legend_handles_labels()\n",
    "    fig.legend(handles, labels, loc='upper right')\n",
    "    fig.tight_layout()\n",
    "    if save:\n",
    "        fig.savefig('figures/airbnb_gam.png', bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "06fb0315",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 809x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "beta_hat_lm = np.mean(np.array([beta_hat_lm_s1, beta_hat_lm_s2]), axis = 0)\n",
    "beta_hat_lm_sd = np.mean(np.array([beta_hat_lm_s1_sd, beta_hat_lm_s2_sd]), axis = 0)\n",
    "\n",
    "plot_lm(beta_hat_lm, beta_hat_lm_sd)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "b8bb3e4e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 970.8x600 with 6 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "beta_hat_gam = np.mean(np.array(([beta_hat_gam1_s1, beta_hat_gam1_s2])), axis = 0)\n",
    "\n",
    "plot_gam(beta_hat_gam)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "16ba100e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 970.8x600 with 6 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_gam_lm(beta_hat_gam, beta_hat_lm, beta_hat_lm_sd)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "440e9bb5",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.10.5 64-bit",
   "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": "3.10.5"
  },
  "vscode": {
   "interpreter": {
    "hash": "369f2c481f4da34e4445cda3fffd2e751bd1c4d706f27375911949ba6bb62e1c"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}