{
"cells": [
{
"cell_type": "markdown",
"id": "4e20bf01",
"metadata": {},
"source": [
"# SARIMAX"
]
},
{
"cell_type": "code",
"execution_count": 1,
"id": "7b7b5183",
"metadata": {},
"outputs": [],
"source": [
"import os\n",
"os.chdir(\"../../\")\n",
"\n",
"import numpy as np\n",
"import pandas as pd\n",
"\n",
"import matplotlib.pyplot as plt\n",
"import seaborn as sns\n",
"sns.set_style(\"whitegrid\")\n",
"\n",
"from statsmodels.tsa.seasonal import seasonal_decompose, STL\n",
"from statsmodels.tsa.statespace.sarimax import SARIMAX\n",
"from statsmodels.tsa.arima.model import ARIMA\n",
"import pmdarima as pm\n",
"from pmdarima import model_selection\n",
"from pmdarima import auto_arima\n",
"from scripts.python.tsa.ts_utils import * \n",
"from scripts.python.tsa.utsmodel import * \n",
"from scripts.python.tsa.ts_eval import *\n",
"\n",
"import warnings\n",
"warnings.filterwarnings('ignore')"
]
},
{
"cell_type": "code",
"execution_count": 2,
"id": "643e36df",
"metadata": {
"scrolled": false,
"tags": [
"hide-input"
]
},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" date | \n",
" total | \n",
" stringency_index | \n",
" covid | \n",
" solomon_islands_flights | \n",
" solomon_islands_hotel | \n",
" solomon_islands_travel | \n",
"
\n",
" \n",
" \n",
" \n",
" | 0 | \n",
" 2009-01-01 | \n",
" 1602 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 0.000000 | \n",
" 0.0 | \n",
" 0.000000 | \n",
"
\n",
" \n",
" | 1 | \n",
" 2009-02-01 | \n",
" 1422 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 0.484667 | \n",
" 0.0 | \n",
" 0.203672 | \n",
"
\n",
" \n",
" | 2 | \n",
" 2009-03-01 | \n",
" 1249 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 0.000000 | \n",
" 0.0 | \n",
" 0.000000 | \n",
"
\n",
" \n",
" | 3 | \n",
" 2009-04-01 | \n",
" 1499 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 0.000000 | \n",
" 0.0 | \n",
" 0.000000 | \n",
"
\n",
" \n",
" | 4 | \n",
" 2009-05-01 | \n",
" 1393 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 0.000000 | \n",
" 0.0 | \n",
" 0.478020 | \n",
"
\n",
" \n",
"
\n",
"
"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"STARTED TO CONDUCT GRID SEARCHING FOR SOLOMON_ISLANDS \n",
"\n",
"STARTED TO CONDUCT CROSS-VALIDATION FOR SOLOMON_ISLANDS \n",
"\n",
"Best Models from Cross-validation is ARIMA(2,0,2)(0,1,1)[12] intercept \n",
"\n"
]
},
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" MSE | \n",
" RMSE | \n",
" MAE | \n",
" SMAPE | \n",
"
\n",
" \n",
" \n",
" \n",
" | naive | \n",
" 130146.048951 | \n",
" 360.757604 | \n",
" 273.083916 | \n",
" 19.605133 | \n",
"
\n",
" \n",
" | mean | \n",
" 373388.932822 | \n",
" 611.055589 | \n",
" 428.780575 | \n",
" 28.510498 | \n",
"
\n",
" \n",
" | seasonal naive | \n",
" 719201.125000 | \n",
" 848.057265 | \n",
" 537.375000 | \n",
" 42.210038 | \n",
"
\n",
" \n",
" | SARIMAX | \n",
" 81169.100331 | \n",
" 284.901914 | \n",
" 201.117080 | \n",
" 14.616283 | \n",
"
\n",
" \n",
"
\n",
"
\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" date | \n",
" japan | \n",
" south_korea | \n",
" taiwan | \n",
" china | \n",
" usa/canada | \n",
" europe | \n",
" others | \n",
" total | \n",
" stringency_index | \n",
" covid | \n",
" palau_flights | \n",
" palau_hotel | \n",
" palau_travel | \n",
"
\n",
" \n",
" \n",
" \n",
" | 0 | \n",
" 2007-06-01 | \n",
" 856.0 | \n",
" 1291.0 | \n",
" 3245.0 | \n",
" 86.0 | \n",
" 669.0 | \n",
" 99.0 | \n",
" 463.0 | \n",
" 6709.0 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 1.068426 | \n",
" 8.334537 | \n",
" 0.320004 | \n",
"
\n",
" \n",
" | 1 | \n",
" 2007-07-01 | \n",
" 2119.0 | \n",
" 1366.0 | \n",
" 3269.0 | \n",
" 33.0 | \n",
" 653.0 | \n",
" 144.0 | \n",
" 437.0 | \n",
" 8021.0 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 0.000000 | \n",
" 9.667343 | \n",
" 0.952482 | \n",
"
\n",
" \n",
" | 2 | \n",
" 2007-08-01 | \n",
" 3476.0 | \n",
" 1354.0 | \n",
" 3046.0 | \n",
" 46.0 | \n",
" 580.0 | \n",
" 256.0 | \n",
" 438.0 | \n",
" 9196.0 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 0.704914 | \n",
" 6.906282 | \n",
" 0.453967 | \n",
"
\n",
" \n",
" | 3 | \n",
" 2007-09-01 | \n",
" 3022.0 | \n",
" 910.0 | \n",
" 2497.0 | \n",
" 61.0 | \n",
" 559.0 | \n",
" 145.0 | \n",
" 401.0 | \n",
" 7595.0 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 0.798284 | \n",
" 3.591027 | \n",
" 0.000000 | \n",
"
\n",
" \n",
" | 4 | \n",
" 2007-10-01 | \n",
" 1807.0 | \n",
" 1082.0 | \n",
" 2298.0 | \n",
" 49.0 | \n",
" 774.0 | \n",
" 390.0 | \n",
" 395.0 | \n",
" 6795.0 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 0.431326 | \n",
" 4.111283 | \n",
" 0.000000 | \n",
"
\n",
" \n",
"
\n",
"
"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"STARTED TO CONDUCT GRID SEARCHING FOR PALAU \n",
"\n",
"STARTED TO CONDUCT CROSS-VALIDATION FOR PALAU \n",
"\n",
"Best Models from Cross-validation is ARIMA(0,1,1)(0,1,1)[12] intercept \n",
"\n"
]
},
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" MSE | \n",
" RMSE | \n",
" MAE | \n",
" SMAPE | \n",
"
\n",
" \n",
" \n",
" \n",
" | naive | \n",
" 2.096767e+06 | \n",
" 1448.021907 | \n",
" 1092.759563 | \n",
" 23.272616 | \n",
"
\n",
" \n",
" | mean | \n",
" 1.639656e+07 | \n",
" 4049.266859 | \n",
" 3178.648757 | \n",
" 51.716972 | \n",
"
\n",
" \n",
" | seasonal naive | \n",
" 1.123704e+07 | \n",
" 3352.169852 | \n",
" 2405.935201 | \n",
" 56.509474 | \n",
"
\n",
" \n",
" | SARIMAX | \n",
" 1.826282e+06 | \n",
" 1351.399848 | \n",
" 947.467373 | \n",
" 21.701883 | \n",
"
\n",
" \n",
"
\n",
"
\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" date | \n",
" total | \n",
" american_samoa | \n",
" australia | \n",
" europe | \n",
" new_zealand | \n",
" usa | \n",
" other_countries | \n",
" cruise_ships | \n",
" stringency_index | \n",
" covid | \n",
" samoa_flights | \n",
" samoa_hotel | \n",
" samoa_travel | \n",
"
\n",
" \n",
" \n",
" \n",
" | 0 | \n",
" 2002-08-01 | \n",
" 8260.0 | \n",
" 2877.0 | \n",
" 1115.0 | \n",
" 429.0 | \n",
" 2035.0 | \n",
" 933.0 | \n",
" 871.0 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 0.0 | \n",
"
\n",
" \n",
" | 1 | \n",
" 2002-09-01 | \n",
" 6708.0 | \n",
" 2119.0 | \n",
" 972.0 | \n",
" 372.0 | \n",
" 1879.0 | \n",
" 612.0 | \n",
" 754.0 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 0.0 | \n",
"
\n",
" \n",
" | 2 | \n",
" 2002-10-01 | \n",
" 5737.0 | \n",
" 1892.0 | \n",
" 737.0 | \n",
" 468.0 | \n",
" 1492.0 | \n",
" 577.0 | \n",
" 571.0 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 0.0 | \n",
"
\n",
" \n",
" | 3 | \n",
" 2002-11-01 | \n",
" 6653.0 | \n",
" 2530.0 | \n",
" 864.0 | \n",
" 393.0 | \n",
" 1675.0 | \n",
" 483.0 | \n",
" 708.0 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 0.0 | \n",
"
\n",
" \n",
" | 4 | \n",
" 2002-12-01 | \n",
" 13042.0 | \n",
" 4678.0 | \n",
" 1918.0 | \n",
" 272.0 | \n",
" 4401.0 | \n",
" 1007.0 | \n",
" 766.0 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 0.0 | \n",
"
\n",
" \n",
"
\n",
"
"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"STARTED TO CONDUCT GRID SEARCHING FOR SAMOA \n",
"\n",
"STARTED TO CONDUCT CROSS-VALIDATION FOR SAMOA \n",
"\n",
"Best Models from Cross-validation is ARIMA(2,1,1)(0,1,1)[12] intercept \n",
"\n"
]
},
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" MSE | \n",
" RMSE | \n",
" MAE | \n",
" SMAPE | \n",
"
\n",
" \n",
" \n",
" \n",
" | naive | \n",
" 1.180276e+07 | \n",
" 3435.514859 | \n",
" 2313.560166 | \n",
" 23.936179 | \n",
"
\n",
" \n",
" | mean | \n",
" 2.193805e+07 | \n",
" 4683.807205 | \n",
" 3367.534731 | \n",
" 43.918693 | \n",
"
\n",
" \n",
" | seasonal naive | \n",
" 1.643507e+07 | \n",
" 4054.018920 | \n",
" 2003.082645 | \n",
" 31.466659 | \n",
"
\n",
" \n",
" | SARIMAX | \n",
" 2.952049e+06 | \n",
" 1718.152711 | \n",
" 1020.766988 | \n",
" 33.573105 | \n",
"
\n",
" \n",
"
\n",
"
\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" date | \n",
" australia | \n",
" new_zealand | \n",
" new_caledonia | \n",
" other_pic | \n",
" europe | \n",
" north_america | \n",
" japan | \n",
" china | \n",
" other_countries | \n",
" total | \n",
" fileyear | \n",
" not_stated | \n",
" stringency_index | \n",
" covid | \n",
" vanuatu_flights | \n",
" vanuatu_hotel | \n",
" vanuatu_travel | \n",
"
\n",
" \n",
" \n",
" \n",
" | 0 | \n",
" 2003-12-01 | \n",
" 2938.0 | \n",
" 580.0 | \n",
" 646.0 | \n",
" 136.0 | \n",
" 201.0 | \n",
" 88.0 | \n",
" 42.0 | \n",
" 0.0 | \n",
" 103.0 | \n",
" 4734.0 | \n",
" 2007.0 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 0.000000 | \n",
" 0.000000 | \n",
" 0.000000 | \n",
"
\n",
" \n",
" | 1 | \n",
" 2004-01-01 | \n",
" 2864.0 | \n",
" 439.0 | \n",
" 717.0 | \n",
" 122.0 | \n",
" 186.0 | \n",
" 67.0 | \n",
" 36.0 | \n",
" 0.0 | \n",
" 61.0 | \n",
" 4492.0 | \n",
" 2005.0 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 2.232933 | \n",
" 3.102047 | \n",
" 2.453441 | \n",
"
\n",
" \n",
" | 2 | \n",
" 2004-02-01 | \n",
" 2158.0 | \n",
" 194.0 | \n",
" 416.0 | \n",
" 282.0 | \n",
" 229.0 | \n",
" 82.0 | \n",
" 34.0 | \n",
" 0.0 | \n",
" 93.0 | \n",
" 3488.0 | \n",
" 2005.0 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 0.000000 | \n",
" 0.000000 | \n",
" 0.000000 | \n",
"
\n",
" \n",
" | 3 | \n",
" 2004-03-01 | \n",
" 2733.0 | \n",
" 420.0 | \n",
" 255.0 | \n",
" 160.0 | \n",
" 195.0 | \n",
" 115.0 | \n",
" 53.0 | \n",
" 0.0 | \n",
" 83.0 | \n",
" 4014.0 | \n",
" 2005.0 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 0.000000 | \n",
" 0.000000 | \n",
" 0.000000 | \n",
"
\n",
" \n",
" | 4 | \n",
" 2004-04-01 | \n",
" 2377.0 | \n",
" 438.0 | \n",
" 469.0 | \n",
" 206.0 | \n",
" 225.0 | \n",
" 115.0 | \n",
" 45.0 | \n",
" 0.0 | \n",
" 80.0 | \n",
" 3955.0 | \n",
" 2005.0 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 0.000000 | \n",
" 0.000000 | \n",
" 1.694580 | \n",
"
\n",
" \n",
"
\n",
"
"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"STARTED TO CONDUCT GRID SEARCHING FOR VANUATU \n",
"\n",
"STARTED TO CONDUCT CROSS-VALIDATION FOR VANUATU \n",
"\n",
"Best Models from Cross-validation is ARIMA(1,0,1)(2,1,2)[12] intercept \n",
"\n"
]
},
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" MSE | \n",
" RMSE | \n",
" MAE | \n",
" SMAPE | \n",
"
\n",
" \n",
" \n",
" \n",
" | naive | \n",
" 2.915180e+06 | \n",
" 1707.389827 | \n",
" 1292.829596 | \n",
" 20.537447 | \n",
"
\n",
" \n",
" | mean | \n",
" 1.125218e+07 | \n",
" 3354.426412 | \n",
" 2628.520727 | \n",
" 47.767194 | \n",
"
\n",
" \n",
" | seasonal naive | \n",
" 8.396013e+06 | \n",
" 2897.587484 | \n",
" 1568.424107 | \n",
" 34.617696 | \n",
"
\n",
" \n",
" | SARIMAX | \n",
" 1.238855e+06 | \n",
" 1113.038657 | \n",
" 795.607455 | \n",
" 35.718702 | \n",
"
\n",
" \n",
"
\n",
"
\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" date | \n",
" yacht | \n",
" all | \n",
" total | \n",
" ship | \n",
" stringency_index | \n",
" covid | \n",
" tonga_flights | \n",
" tonga_hotel | \n",
" tonga_travel | \n",
"
\n",
" \n",
" \n",
" \n",
" | 0 | \n",
" 2010-01-01 | \n",
" 4 | \n",
" 3808 | \n",
" 3158 | \n",
" 646 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 1.118195 | \n",
" 1.296399 | \n",
" 0.212187 | \n",
"
\n",
" \n",
" | 1 | \n",
" 2010-02-01 | \n",
" 5 | \n",
" 2384 | \n",
" 2379 | \n",
" 0 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 0.811451 | \n",
" 1.013375 | \n",
" 0.480398 | \n",
"
\n",
" \n",
" | 2 | \n",
" 2010-03-01 | \n",
" 5 | \n",
" 3992 | \n",
" 3134 | \n",
" 853 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 0.872967 | \n",
" 1.359485 | \n",
" 0.233651 | \n",
"
\n",
" \n",
" | 3 | \n",
" 2010-04-01 | \n",
" 30 | \n",
" 5650 | \n",
" 2818 | \n",
" 2802 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 0.583649 | \n",
" 1.142021 | \n",
" 0.400232 | \n",
"
\n",
" \n",
" | 4 | \n",
" 2010-05-01 | \n",
" 177 | \n",
" 9150 | \n",
" 3670 | \n",
" 5303 | \n",
" 0.0 | \n",
" 0.0 | \n",
" 0.906579 | \n",
" 1.746620 | \n",
" 0.749327 | \n",
"
\n",
" \n",
"
\n",
"
"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"STARTED TO CONDUCT GRID SEARCHING FOR TONGA \n",
"\n",
"STARTED TO CONDUCT CROSS-VALIDATION FOR TONGA \n",
"\n",
"Best Models from Cross-validation is ARIMA(0,0,1)(0,1,1)[12] intercept \n",
"\n"
]
},
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" MSE | \n",
" RMSE | \n",
" MAE | \n",
" SMAPE | \n",
"
\n",
" \n",
" \n",
" \n",
" | naive | \n",
" 2.543522e+06 | \n",
" 1594.842251 | \n",
" 1043.874126 | \n",
" 35.445486 | \n",
"
\n",
" \n",
" | mean | \n",
" 4.529700e+06 | \n",
" 2128.309258 | \n",
" 1637.978395 | \n",
" 53.528230 | \n",
"
\n",
" \n",
" | seasonal naive | \n",
" 4.760284e+06 | \n",
" 2181.807619 | \n",
" 1256.819444 | \n",
" 53.730230 | \n",
"
\n",
" \n",
" | SARIMAX | \n",
" 7.153640e+05 | \n",
" 845.791954 | \n",
" 556.793463 | \n",
" 30.118200 | \n",
"
\n",
" \n",
"
\n",
"