Data Modeling - ARIMA.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "ac2009ad",
   "metadata": {},
   "source": [
    "# Drilling Activity Prediction: Data Modeling - ARIMA\n",
    "\n",
    "- Import dataset and resample to monthly data\n",
    "- Use oil dataset to predict oil rig count:\n",
    "    - Find what hyperparameters to be used in the model (p,d,q)\n",
    "    - Train/test split\n",
    "    - Create ARIMA model and fit with data\n",
    "    - Generate Forecast and compare against Test set\n",
    "    - Evaluation Measures\n",
    "    - Refit the model with 100% of data and make REAL Forecasting    \n",
    "    \n",
    "    \n",
    " - Create ARIMA model with Gas dataframe to predict Gas Rig Count"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "ca40b7d3",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "# Load specific forecasting tools\n",
    "from statsmodels.tsa.arima.model import ARIMA,ARIMAResults\n",
    "from statsmodels.tsa.statespace.sarimax import SARIMAX\n",
    "from statsmodels.tsa.seasonal import seasonal_decompose\n",
    "from statsmodels.tools.eval_measures import rmse, mse, meanabs\n",
    "from sklearn.model_selection import TimeSeriesSplit\n",
    "import time\n",
    "from pmdarima import auto_arima\n",
    "from pylab import rcParams\n",
    "\n",
    "# Ignore harmless warnings\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")\n",
    "\n",
    "# Size of all plots\n",
    "rcParams['figure.figsize'] = 6,6"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "d0d597a0",
   "metadata": {},
   "outputs": [],
   "source": [
    "dfoil = pd.read_csv('dfoil.csv',index_col=0,parse_dates=True)\n",
    "dfoil = dfoil.resample('M').mean()\n",
    "dfoil.index.freq = 'M'\n",
    "\n",
    "\n",
    "dfgas = pd.read_csv('dfgas.csv',index_col=0,parse_dates=True)\n",
    "dfgas = dfgas.resample('M').mean()\n",
    "dfgas.index.freq = 'M'"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "79a1d56b",
   "metadata": {},
   "source": [
    "Before we can apply an ARIMA forecasting model, we need to review the components of one.<br>\n",
    "ARIMA, or Autoregressive Independent Moving Average is actually a combination of 3 models:\n",
    "* <strong>AR(p)</strong> Autoregression - a regression model that utilizes the dependent relationship between a current observation and observations over a previous period.\n",
    "* <strong>I(d)</strong> Integration - uses differencing of observations (subtracting an observation from an observation at the previous time step) in order to make the time series stationary\n",
    "* <strong>MA(q)</strong> Moving Average - a model that uses the dependency between an observation and a residual error from a moving average model applied to lagged observations.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "3665bdee",
   "metadata": {},
   "outputs": [],
   "source": [
    "nobs = 12"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "c37dfe8a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Performing stepwise search to minimize aic\n",
      " ARIMA(0,1,0)(1,0,1)[12] intercept   : AIC=2965.143, Time=0.48 sec\n",
      " ARIMA(0,1,0)(0,0,0)[12] intercept   : AIC=2964.110, Time=0.02 sec\n",
      " ARIMA(1,1,0)(1,0,0)[12] intercept   : AIC=2595.587, Time=0.44 sec\n",
      " ARIMA(0,1,1)(0,0,1)[12] intercept   : AIC=inf, Time=0.87 sec\n",
      " ARIMA(0,1,0)(0,0,0)[12]             : AIC=2962.119, Time=0.02 sec\n",
      " ARIMA(1,1,0)(0,0,0)[12] intercept   : AIC=2595.900, Time=0.08 sec\n",
      " ARIMA(1,1,0)(2,0,0)[12] intercept   : AIC=2596.893, Time=1.05 sec\n",
      " ARIMA(1,1,0)(1,0,1)[12] intercept   : AIC=2597.400, Time=1.05 sec\n",
      " ARIMA(1,1,0)(0,0,1)[12] intercept   : AIC=2595.435, Time=0.51 sec\n",
      " ARIMA(1,1,0)(0,0,2)[12] intercept   : AIC=2597.316, Time=1.23 sec\n",
      " ARIMA(1,1,0)(1,0,2)[12] intercept   : AIC=2599.051, Time=1.40 sec\n",
      " ARIMA(0,1,0)(0,0,1)[12] intercept   : AIC=2963.810, Time=0.28 sec\n",
      " ARIMA(2,1,0)(0,0,1)[12] intercept   : AIC=2496.812, Time=0.59 sec\n",
      " ARIMA(2,1,0)(0,0,0)[12] intercept   : AIC=2498.442, Time=0.21 sec\n",
      " ARIMA(2,1,0)(1,0,1)[12] intercept   : AIC=2498.811, Time=1.01 sec\n",
      " ARIMA(2,1,0)(0,0,2)[12] intercept   : AIC=2498.808, Time=1.45 sec\n",
      " ARIMA(2,1,0)(1,0,0)[12] intercept   : AIC=2496.895, Time=0.63 sec\n",
      " ARIMA(2,1,0)(1,0,2)[12] intercept   : AIC=2500.765, Time=3.09 sec\n",
      " ARIMA(3,1,0)(0,0,1)[12] intercept   : AIC=2447.724, Time=0.67 sec\n",
      " ARIMA(3,1,0)(0,0,0)[12] intercept   : AIC=2447.856, Time=0.30 sec\n",
      " ARIMA(3,1,0)(1,0,1)[12] intercept   : AIC=2449.677, Time=1.29 sec\n",
      " ARIMA(3,1,0)(0,0,2)[12] intercept   : AIC=2449.530, Time=1.89 sec\n",
      " ARIMA(3,1,0)(1,0,0)[12] intercept   : AIC=2447.878, Time=0.72 sec\n",
      " ARIMA(3,1,0)(1,0,2)[12] intercept   : AIC=2451.134, Time=2.18 sec\n",
      " ARIMA(4,1,0)(0,0,1)[12] intercept   : AIC=2411.782, Time=0.72 sec\n",
      " ARIMA(4,1,0)(0,0,0)[12] intercept   : AIC=2410.281, Time=0.38 sec\n",
      " ARIMA(4,1,0)(1,0,0)[12] intercept   : AIC=2411.853, Time=0.92 sec\n",
      " ARIMA(4,1,0)(1,0,1)[12] intercept   : AIC=2413.593, Time=1.10 sec\n",
      " ARIMA(5,1,0)(0,0,0)[12] intercept   : AIC=2393.976, Time=0.36 sec\n",
      " ARIMA(5,1,0)(1,0,0)[12] intercept   : AIC=2395.519, Time=0.79 sec\n",
      " ARIMA(5,1,0)(0,0,1)[12] intercept   : AIC=2395.429, Time=0.66 sec\n",
      " ARIMA(5,1,0)(1,0,1)[12] intercept   : AIC=2397.160, Time=1.27 sec\n",
      " ARIMA(6,1,0)(0,0,0)[12] intercept   : AIC=2394.111, Time=0.48 sec\n",
      " ARIMA(5,1,1)(0,0,0)[12] intercept   : AIC=2359.444, Time=0.96 sec\n",
      " ARIMA(5,1,1)(1,0,0)[12] intercept   : AIC=2360.415, Time=1.89 sec\n",
      " ARIMA(5,1,1)(0,0,1)[12] intercept   : AIC=2360.250, Time=1.68 sec\n",
      " ARIMA(5,1,1)(1,0,1)[12] intercept   : AIC=2362.050, Time=3.35 sec\n",
      " ARIMA(4,1,1)(0,0,0)[12] intercept   : AIC=2358.304, Time=0.73 sec\n",
      " ARIMA(4,1,1)(1,0,0)[12] intercept   : AIC=2359.153, Time=1.59 sec\n",
      " ARIMA(4,1,1)(0,0,1)[12] intercept   : AIC=2358.973, Time=1.54 sec\n",
      " ARIMA(4,1,1)(1,0,1)[12] intercept   : AIC=2360.779, Time=2.19 sec\n",
      " ARIMA(3,1,1)(0,0,0)[12] intercept   : AIC=2356.949, Time=0.35 sec\n",
      " ARIMA(3,1,1)(1,0,0)[12] intercept   : AIC=2357.686, Time=1.23 sec\n",
      " ARIMA(3,1,1)(0,0,1)[12] intercept   : AIC=2357.502, Time=0.91 sec\n",
      " ARIMA(3,1,1)(1,0,1)[12] intercept   : AIC=2359.325, Time=1.71 sec\n",
      " ARIMA(2,1,1)(0,0,0)[12] intercept   : AIC=2359.073, Time=0.42 sec\n",
      " ARIMA(3,1,2)(0,0,0)[12] intercept   : AIC=2358.637, Time=0.65 sec\n",
      " ARIMA(2,1,2)(0,0,0)[12] intercept   : AIC=2357.181, Time=0.64 sec\n",
      " ARIMA(4,1,2)(0,0,0)[12] intercept   : AIC=2360.082, Time=0.97 sec\n",
      " ARIMA(3,1,1)(0,0,0)[12]             : AIC=2354.952, Time=0.23 sec\n",
      " ARIMA(3,1,1)(1,0,0)[12]             : AIC=2355.690, Time=0.68 sec\n",
      " ARIMA(3,1,1)(0,0,1)[12]             : AIC=2355.504, Time=0.65 sec\n",
      " ARIMA(3,1,1)(1,0,1)[12]             : AIC=2357.329, Time=0.89 sec\n",
      " ARIMA(2,1,1)(0,0,0)[12]             : AIC=2357.077, Time=0.18 sec\n",
      " ARIMA(3,1,0)(0,0,0)[12]             : AIC=2445.862, Time=0.14 sec\n",
      " ARIMA(4,1,1)(0,0,0)[12]             : AIC=2356.307, Time=0.37 sec\n",
      " ARIMA(3,1,2)(0,0,0)[12]             : AIC=2356.639, Time=0.50 sec\n",
      " ARIMA(2,1,0)(0,0,0)[12]             : AIC=2496.447, Time=0.08 sec\n",
      " ARIMA(2,1,2)(0,0,0)[12]             : AIC=2355.184, Time=0.32 sec\n",
      " ARIMA(4,1,0)(0,0,0)[12]             : AIC=2408.283, Time=0.16 sec\n",
      " ARIMA(4,1,2)(0,0,0)[12]             : AIC=2358.085, Time=0.52 sec\n",
      "\n",
      "Best model:  ARIMA(3,1,1)(0,0,0)[12]          \n",
      "Total fit time: 53.717 seconds\n"
     ]
    }
   ],
   "source": [
    "stepwise_fit = auto_arima(dfoil['oilrig_count'], start_p=0, start_q=0,\n",
    "                          max_p=6, max_q=3, m=nobs,\n",
    "                          seasonal=True,\n",
    "                          d=None, trace=True,\n",
    "                          error_action='ignore',   # we don't want to know if an order does not work\n",
    "                          suppress_warnings=True,  # we don't want convergence warnings\n",
    "                          stepwise=True)           # set to stepwise\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2fc79725",
   "metadata": {},
   "source": [
    "It looks like the order (3,1,1) is recommended for oil df."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "09708753",
   "metadata": {},
   "source": [
    "#### Train/test split\n",
    "\n",
    "- We will evaluate the model forecasting 12 months ahead (1 year)\n",
    "- Train test split. Test df will have 12 data points"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "886e00fb",
   "metadata": {},
   "outputs": [],
   "source": [
    "train = dfoil.iloc[:len(dfoil)-nobs]\n",
    "test = dfoil.iloc[len(dfoil)-nobs:]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d76588b6",
   "metadata": {},
   "source": [
    "#### Fit ARIMA model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "7a7a625a",
   "metadata": {},
   "outputs": [],
   "source": [
    "oil_model = ARIMA(train['oilrig_count'],order=(3,1,1))\n",
    "oil_results = oil_model.fit()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "267712cf",
   "metadata": {},
   "source": [
    "#### Generate Forecast values and plot results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "e8b42149",
   "metadata": {},
   "outputs": [],
   "source": [
    "#12 months forecast\n",
    "start=len(train)\n",
    "end=len(train)+len(test)-1\n",
    "oil_predictions = oil_results.predict(start=start, end=end, dynamic=False, typ='levels').rename('Oil Rig Count ARIMA(3,1,1) Predictions')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "a3236518",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot predictions against known values\n",
    "title = 'Oil Rig Count Predictions Vs Test Set - ARIMA'\n",
    "ylabel='Oil Rig Count'\n",
    "xlabel=''\n",
    "\n",
    "ax = dfoil['oilrig_count'].loc['2017':].plot(legend=True,title=title)\n",
    "oil_predictions.plot(legend=True)\n",
    "ax.autoscale(axis='x',tight=True)\n",
    "ax.set(xlabel=xlabel, ylabel=ylabel);"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1865d752",
   "metadata": {},
   "source": [
    "#### Evaluation Metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "b2002404",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Oil Rig ARIMA(3,1,1) RMSE: 66.322\n",
      "Oil Rig ARIMA(3,1,1) MSE: 4398.668\n",
      "Oil Rig ARIMA(3,1,1) MAE: 55.252\n",
      "\n",
      "Oil Rig Mean Value: 345.055\n",
      "Oil Rig std Value: 54.791\n",
      "Oil Rig Percent Change: 19.221%\n"
     ]
    }
   ],
   "source": [
    "RMSE1 = rmse(dfoil['oilrig_count'].loc[test.index.min():],oil_predictions)\n",
    "MSE1 = mse(dfoil['oilrig_count'].loc[test.index.min():],oil_predictions)\n",
    "MAE1 = meanabs(dfoil['oilrig_count'].loc[test.index.min():],oil_predictions)\n",
    "\n",
    "print(f\"\"\"Oil Rig ARIMA(3,1,1) RMSE: {RMSE1:.3f}\n",
    "Oil Rig ARIMA(3,1,1) MSE: {MSE1:.3f}\n",
    "Oil Rig ARIMA(3,1,1) MAE: {MAE1:.3f}\\n\n",
    "Oil Rig Mean Value: {dfoil['oilrig_count'][-nobs:].mean():.3f}\n",
    "Oil Rig std Value: {dfoil['oilrig_count'][-nobs:].std():.3f}\n",
    "Oil Rig Percent Change: {RMSE1/dfoil['oilrig_count'][-nobs:].mean()*100:.3f}%\"\"\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "f0fe5f12",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TRAIN: 178 TEST: 12\n",
      "TRAIN: 190 TEST: 12\n",
      "TRAIN: 202 TEST: 12\n",
      "TRAIN: 214 TEST: 12\n",
      "TRAIN: 226 TEST: 12\n",
      "TRAIN: 238 TEST: 12\n",
      "TRAIN: 250 TEST: 12\n",
      "TRAIN: 262 TEST: 12\n",
      "TRAIN: 274 TEST: 12\n",
      "TRAIN: 286 TEST: 12\n",
      "\n",
      "10-fold Cross Validation Oil Rig\n",
      "RMSE: 195.867\n",
      "MSE: 77269.604\n",
      "MAE: 170.945\n",
      "\n",
      "Oil Rig Mean Value: 345.055\n",
      "Oil Rig std Value: 54.791\n",
      "Oil Rig Percent Change: 56.764%\n"
     ]
    }
   ],
   "source": [
    "tscv = TimeSeriesSplit(n_splits=10, test_size=12)\n",
    "RMSE_oil = []\n",
    "MSE_oil = []\n",
    "MAE_oil = []\n",
    "for train_index, test_index in tscv.split(dfoil['oilrig_count']):\n",
    "    print(\"TRAIN:\", len(train_index), \"TEST:\", len(test_index))\n",
    "    train, test = dfoil['oilrig_count'][train_index], dfoil['oilrig_count'][test_index]\n",
    "    oil_model = ARIMA(train,order=(3,1,1))\n",
    "    oil_results = oil_model.fit()\n",
    "    start=len(train)\n",
    "    end=len(train)+len(test)-1\n",
    "    oil_predictions = oil_results.predict(start=start, end=end, dynamic=False, typ='levels')   \n",
    "    RMSE_oil.append('{:.3f}'.format(rmse(dfoil.loc[dfoil.index[test_index],'oilrig_count'],oil_predictions)))\n",
    "    MSE_oil.append('{:.3f}'.format(mse(dfoil.loc[dfoil.index[test_index],'oilrig_count'],oil_predictions)))\n",
    "    MAE_oil.append('{:.3f}'.format(meanabs(dfoil.loc[dfoil.index[test_index],'oilrig_count'],oil_predictions)))\n",
    "\n",
    "RMSE_oil = np.array(RMSE_oil).astype('float')\n",
    "MSE_oil = np.array(MSE_oil).astype('float')\n",
    "MAE_oil = np.array(MAE_oil).astype('float')\n",
    "print(f\"\"\"\\n10-fold Cross Validation Oil Rig\n",
    "RMSE: {RMSE_oil.mean():.3f}\n",
    "MSE: {MSE_oil.mean():.3f}\n",
    "MAE: {MAE_oil.mean():.3f}\\n\n",
    "Oil Rig Mean Value: {dfoil['oilrig_count'][-nobs:].mean():.3f}\n",
    "Oil Rig std Value: {dfoil['oilrig_count'][-nobs:].std():.3f}\n",
    "Oil Rig Percent Change: {RMSE_oil.mean()/dfoil['oilrig_count'][-nobs:].mean()*100:.3f}%\"\"\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "10434968",
   "metadata": {},
   "source": [
    "#### Fit Model with 100% of Data and forecast Real Future"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "6ce8889e",
   "metadata": {},
   "outputs": [],
   "source": [
    "startTime = time.time()\n",
    "oil_model = ARIMA(dfoil['oilrig_count'],order=(3,1,1))\n",
    "oil_results = oil_model.fit()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "3d80d758",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Execution time in seconds: 0.2573118209838867\n"
     ]
    }
   ],
   "source": [
    "start=len(dfoil)\n",
    "end=len(dfoil)+nobs\n",
    "oil_predictions = oil_results.predict(start=start, end=end, dynamic=False, typ='levels').rename('Oil Rig Count ARIMA(3,1,1) Predictions')\n",
    "executionTime = (time.time() - startTime)\n",
    "print('Execution time in seconds: ' + str(executionTime))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "70322d26",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot predictions against known values\n",
    "title = 'Oil Rig Count REAL Predictions - ARIMA'\n",
    "ylabel='Oil Rig Count'\n",
    "xlabel=''\n",
    "\n",
    "ax = dfoil['oilrig_count'].loc['2017':].plot(legend=True,title=title)\n",
    "oil_predictions.plot(legend=True)\n",
    "ax.autoscale(axis='x',tight=True)\n",
    "ax.set(xlabel=xlabel, ylabel=ylabel);"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1260cfb3",
   "metadata": {},
   "source": [
    "#### Create model with Gas Rig Counts"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "0371bf2f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Performing stepwise search to minimize aic\n",
      " ARIMA(0,1,0)(1,0,1)[12] intercept   : AIC=2856.211, Time=0.76 sec\n",
      " ARIMA(0,1,0)(0,0,0)[12] intercept   : AIC=2853.367, Time=0.02 sec\n",
      " ARIMA(1,1,0)(1,0,0)[12] intercept   : AIC=2446.106, Time=0.44 sec\n",
      " ARIMA(0,1,1)(0,0,1)[12] intercept   : AIC=2507.177, Time=0.75 sec\n",
      " ARIMA(0,1,0)(0,0,0)[12]             : AIC=2851.947, Time=0.02 sec\n",
      " ARIMA(1,1,0)(0,0,0)[12] intercept   : AIC=2445.238, Time=0.07 sec\n",
      " ARIMA(1,1,0)(0,0,1)[12] intercept   : AIC=2446.116, Time=0.38 sec\n",
      " ARIMA(1,1,0)(1,0,1)[12] intercept   : AIC=inf, Time=1.29 sec\n",
      " ARIMA(2,1,0)(0,0,0)[12] intercept   : AIC=2388.371, Time=0.20 sec\n",
      " ARIMA(2,1,0)(1,0,0)[12] intercept   : AIC=2389.692, Time=0.65 sec\n",
      " ARIMA(2,1,0)(0,0,1)[12] intercept   : AIC=2389.608, Time=0.57 sec\n",
      " ARIMA(2,1,0)(1,0,1)[12] intercept   : AIC=2391.395, Time=1.11 sec\n",
      " ARIMA(3,1,0)(0,0,0)[12] intercept   : AIC=2380.127, Time=0.29 sec\n",
      " ARIMA(3,1,0)(1,0,0)[12] intercept   : AIC=2381.190, Time=0.67 sec\n",
      " ARIMA(3,1,0)(0,0,1)[12] intercept   : AIC=2381.120, Time=0.67 sec\n",
      " ARIMA(3,1,0)(1,0,1)[12] intercept   : AIC=2383.017, Time=0.91 sec\n",
      " ARIMA(4,1,0)(0,0,0)[12] intercept   : AIC=2328.889, Time=0.36 sec\n",
      " ARIMA(4,1,0)(1,0,0)[12] intercept   : AIC=2329.369, Time=1.06 sec\n",
      " ARIMA(4,1,0)(0,0,1)[12] intercept   : AIC=2329.291, Time=0.83 sec\n",
      " ARIMA(4,1,0)(1,0,1)[12] intercept   : AIC=2331.239, Time=1.31 sec\n",
      " ARIMA(5,1,0)(0,0,0)[12] intercept   : AIC=2318.116, Time=0.43 sec\n",
      " ARIMA(5,1,0)(1,0,0)[12] intercept   : AIC=2319.693, Time=1.27 sec\n",
      " ARIMA(5,1,0)(0,0,1)[12] intercept   : AIC=2319.664, Time=0.91 sec\n",
      " ARIMA(5,1,0)(1,0,1)[12] intercept   : AIC=2321.549, Time=2.39 sec\n",
      " ARIMA(6,1,0)(0,0,0)[12] intercept   : AIC=2308.135, Time=0.67 sec\n",
      " ARIMA(6,1,0)(1,0,0)[12] intercept   : AIC=2309.062, Time=1.77 sec\n",
      " ARIMA(6,1,0)(0,0,1)[12] intercept   : AIC=2309.018, Time=1.17 sec\n",
      " ARIMA(6,1,0)(1,0,1)[12] intercept   : AIC=2310.950, Time=2.76 sec\n",
      " ARIMA(6,1,1)(0,0,0)[12] intercept   : AIC=2289.883, Time=1.22 sec\n",
      " ARIMA(6,1,1)(1,0,0)[12] intercept   : AIC=2291.370, Time=2.52 sec\n",
      " ARIMA(6,1,1)(0,0,1)[12] intercept   : AIC=2291.346, Time=2.09 sec\n",
      " ARIMA(6,1,1)(1,0,1)[12] intercept   : AIC=2293.263, Time=3.33 sec\n",
      " ARIMA(5,1,1)(0,0,0)[12] intercept   : AIC=2287.961, Time=0.80 sec\n",
      " ARIMA(5,1,1)(1,0,0)[12] intercept   : AIC=2289.452, Time=2.00 sec\n",
      " ARIMA(5,1,1)(0,0,1)[12] intercept   : AIC=2289.429, Time=1.62 sec\n",
      " ARIMA(5,1,1)(1,0,1)[12] intercept   : AIC=2291.345, Time=2.71 sec\n",
      " ARIMA(4,1,1)(0,0,0)[12] intercept   : AIC=2287.664, Time=0.41 sec\n",
      " ARIMA(4,1,1)(1,0,0)[12] intercept   : AIC=2289.485, Time=1.03 sec\n",
      " ARIMA(4,1,1)(0,0,1)[12] intercept   : AIC=2289.476, Time=1.08 sec\n",
      " ARIMA(4,1,1)(1,0,1)[12] intercept   : AIC=inf, Time=2.72 sec\n",
      " ARIMA(3,1,1)(0,0,0)[12] intercept   : AIC=2301.503, Time=0.42 sec\n",
      " ARIMA(4,1,2)(0,0,0)[12] intercept   : AIC=2287.907, Time=0.73 sec\n",
      " ARIMA(3,1,2)(0,0,0)[12] intercept   : AIC=2299.119, Time=0.54 sec\n",
      " ARIMA(5,1,2)(0,0,0)[12] intercept   : AIC=2289.861, Time=1.15 sec\n",
      " ARIMA(4,1,1)(0,0,0)[12]             : AIC=2285.772, Time=0.26 sec\n",
      " ARIMA(4,1,1)(1,0,0)[12]             : AIC=2287.582, Time=0.62 sec\n",
      " ARIMA(4,1,1)(0,0,1)[12]             : AIC=2287.573, Time=0.56 sec\n",
      " ARIMA(4,1,1)(1,0,1)[12]             : AIC=inf, Time=2.27 sec\n",
      " ARIMA(3,1,1)(0,0,0)[12]             : AIC=2299.560, Time=0.19 sec\n",
      " ARIMA(4,1,0)(0,0,0)[12]             : AIC=2327.075, Time=0.15 sec\n",
      " ARIMA(5,1,1)(0,0,0)[12]             : AIC=2286.094, Time=0.45 sec\n",
      " ARIMA(4,1,2)(0,0,0)[12]             : AIC=2286.050, Time=0.35 sec\n",
      " ARIMA(3,1,0)(0,0,0)[12]             : AIC=2378.179, Time=0.09 sec\n",
      " ARIMA(3,1,2)(0,0,0)[12]             : AIC=2297.288, Time=0.35 sec\n",
      " ARIMA(5,1,0)(0,0,0)[12]             : AIC=2316.228, Time=0.16 sec\n",
      " ARIMA(5,1,2)(0,0,0)[12]             : AIC=2288.001, Time=0.75 sec\n",
      "\n",
      "Best model:  ARIMA(4,1,1)(0,0,0)[12]          \n",
      "Total fit time: 54.366 seconds\n"
     ]
    }
   ],
   "source": [
    "stepwise_fit = auto_arima(dfgas['gasrig_count'], start_p=0, start_q=0,\n",
    "                          max_p=6, max_q=3, m=nobs,\n",
    "                          seasonal=True,\n",
    "                          d=None, trace=True,\n",
    "                          error_action='ignore',   # we don't want to know if an order does not work\n",
    "                          suppress_warnings=True,  # we don't want convergence warnings\n",
    "                          stepwise=True)           # set to stepwise"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "1ef8c318",
   "metadata": {},
   "outputs": [],
   "source": [
    "train = dfgas.iloc[:len(dfgas)-nobs]\n",
    "test = dfgas.iloc[len(dfgas)-nobs:]\n",
    "gas_model = ARIMA(train['gasrig_count'],order=(4,1,1))\n",
    "gas_results = gas_model.fit()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "31ac159d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "start=len(train)\n",
    "end=len(train)+len(test)-1\n",
    "gas_predictions = gas_results.predict(start=start, end=end, dynamic=False, typ='levels').rename('Gas Rig Count ARIMA(4,1,1) Predictions')\n",
    "# Plot predictions against known values\n",
    "title = 'Gas Rig Count Predictions Vs Test Set - ARIMA'\n",
    "ylabel='Gas Rig Count'\n",
    "xlabel=''\n",
    "\n",
    "ax = dfgas['gasrig_count'].loc['2017':].plot(legend=True,title=title)\n",
    "gas_predictions.plot(legend=True)\n",
    "ax.autoscale(axis='x',tight=True)\n",
    "ax.set(xlabel=xlabel, ylabel=ylabel);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "9eb23afd",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Gas Rig ARIMA(4,1,1) RMSE: 17.085\n",
      "Gas Rig ARIMA(4,1,1) MSE: 291.883\n",
      "Gas Rig ARIMA(4,1,1) MAE: 15.785\n",
      "\n",
      "Gas Rig Mean Value: 94.254\n",
      "Gas Rig std Value: 7.984\n",
      "Gas Rig Percent Change: 18.126%\n"
     ]
    }
   ],
   "source": [
    "RMSE2 = rmse(dfgas['gasrig_count'].loc[test.index.min():],gas_predictions)\n",
    "MSE2 = mse(dfgas['gasrig_count'].loc[test.index.min():],gas_predictions)\n",
    "MAE2 = meanabs(dfgas['gasrig_count'].loc[test.index.min():],gas_predictions)\n",
    "\n",
    "print(f\"\"\"Gas Rig ARIMA(4,1,1) RMSE: {RMSE2:.3f}\n",
    "Gas Rig ARIMA(4,1,1) MSE: {MSE2:.3f}\n",
    "Gas Rig ARIMA(4,1,1) MAE: {MAE2:.3f}\\n\n",
    "Gas Rig Mean Value: {dfgas['gasrig_count'][-nobs:].mean():.3f}\n",
    "Gas Rig std Value: {dfgas['gasrig_count'][-nobs:].std():.3f}\n",
    "Gas Rig Percent Change: {RMSE2/dfgas['gasrig_count'][-nobs:].mean()*100:.3f}%\"\"\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "2f40a753",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TRAIN: 178 TEST: 12\n",
      "TRAIN: 190 TEST: 12\n",
      "TRAIN: 202 TEST: 12\n",
      "TRAIN: 214 TEST: 12\n",
      "TRAIN: 226 TEST: 12\n",
      "TRAIN: 238 TEST: 12\n",
      "TRAIN: 250 TEST: 12\n",
      "TRAIN: 262 TEST: 12\n",
      "TRAIN: 274 TEST: 12\n",
      "TRAIN: 286 TEST: 12\n",
      "\n",
      "10-fold Cross Validation Gas Rig\n",
      "RMSE: 64.389\n",
      "MSE: 8689.194\n",
      "MAE: 57.037\n",
      "\n",
      "Gas Rig Mean Value: 94.254\n",
      "Gas Rig std Value: 7.984\n",
      "Gas Rig Percent Change: 68.314%\n"
     ]
    }
   ],
   "source": [
    "tscv = TimeSeriesSplit(n_splits=10, test_size=12)\n",
    "RMSE_gas = []\n",
    "MSE_gas = []\n",
    "MAE_gas = []\n",
    "for train_index, test_index in tscv.split(dfgas['gasrig_count']):\n",
    "    print(\"TRAIN:\", len(train_index), \"TEST:\", len(test_index))\n",
    "    train, test = dfgas['gasrig_count'][train_index], dfgas['gasrig_count'][test_index]\n",
    "    gas_model = ARIMA(train,order=(3,1,1))\n",
    "    gas_results = gas_model.fit()\n",
    "    start=len(train)\n",
    "    end=len(train)+len(test)-1\n",
    "    gas_predictions = gas_results.predict(start=start, end=end, dynamic=False, typ='levels')   \n",
    "    RMSE_gas.append('{:.3f}'.format(rmse(dfgas.loc[dfgas.index[test_index],'gasrig_count'],gas_predictions)))\n",
    "    MSE_gas.append('{:.3f}'.format(mse(dfgas.loc[dfgas.index[test_index],'gasrig_count'],gas_predictions)))\n",
    "    MAE_gas.append('{:.3f}'.format(meanabs(dfgas.loc[dfgas.index[test_index],'gasrig_count'],gas_predictions)))\n",
    "\n",
    "RMSE_gas = np.array(RMSE_gas).astype('float')\n",
    "MSE_gas = np.array(MSE_gas).astype('float')\n",
    "MAE_gas = np.array(MAE_gas).astype('float')\n",
    "print(f\"\"\"\\n10-fold Cross Validation Gas Rig\n",
    "RMSE: {RMSE_gas.mean():.3f}\n",
    "MSE: {MSE_gas.mean():.3f}\n",
    "MAE: {MAE_gas.mean():.3f}\\n\n",
    "Gas Rig Mean Value: {dfgas['gasrig_count'][-nobs:].mean():.3f}\n",
    "Gas Rig std Value: {dfgas['gasrig_count'][-nobs:].std():.3f}\n",
    "Gas Rig Percent Change: {RMSE_gas.mean()/dfgas['gasrig_count'][-nobs:].mean()*100:.3f}%\"\"\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a6f6c52d",
   "metadata": {},
   "source": [
    "#### Gas Rig Count REAL PREDICTIONS with 100% data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "149020f8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "gas_model = ARIMA(dfgas['gasrig_count'],order=(4,1,1))\n",
    "gas_results = gas_model.fit()\n",
    "start=len(dfgas)\n",
    "end=len(dfgas)+nobs\n",
    "gas_predictions = gas_results.predict(start=start, end=end, dynamic=False, typ='levels').rename('Gas Rig Count ARIMA(4,1,1) Predictions')\n",
    "# Plot predictions against known values\n",
    "title = 'Gas Rig Count REAL Predictions - ARIMA'\n",
    "ylabel='Gas Rig Count'\n",
    "xlabel=''\n",
    "\n",
    "ax = dfgas['gasrig_count'].loc['2017':].plot(legend=True,title=title)\n",
    "gas_predictions.plot(legend=True)\n",
    "ax.autoscale(axis='x',tight=True)\n",
    "ax.set(xlabel=xlabel, ylabel=ylabel);"
   ]
  }
 ],
 "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": "3.9.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}