Simple auto_arima model

This is a simple example of how we can fit an ARIMA model in several lines without knowing anything about our data or optimal hyper parameters.

Out:

Fit ARIMA: order=(2, 1, 2) seasonal_order=(1, 0, 1, 12); AIC=2908.451, BIC=2932.482, Fit time=1.065 seconds
Fit ARIMA: order=(0, 1, 0) seasonal_order=(0, 0, 0, 12); AIC=3049.597, BIC=3055.604, Fit time=0.038 seconds
Fit ARIMA: order=(1, 1, 0) seasonal_order=(1, 0, 0, 12); AIC=2895.934, BIC=2907.950, Fit time=0.304 seconds
Fit ARIMA: order=(0, 1, 1) seasonal_order=(0, 0, 1, 12); AIC=2917.891, BIC=2929.907, Fit time=0.520 seconds
Fit ARIMA: order=(1, 1, 0) seasonal_order=(0, 0, 0, 12); AIC=3039.240, BIC=3048.252, Fit time=0.013 seconds
Fit ARIMA: order=(1, 1, 0) seasonal_order=(2, 0, 0, 12); AIC=2890.929, BIC=2905.949, Fit time=3.189 seconds
Fit ARIMA: order=(1, 1, 0) seasonal_order=(2, 0, 1, 12); AIC=2892.498, BIC=2910.522, Fit time=4.793 seconds
Fit ARIMA: order=(0, 1, 0) seasonal_order=(2, 0, 0, 12); AIC=2906.911, BIC=2918.927, Fit time=2.095 seconds
Fit ARIMA: order=(2, 1, 0) seasonal_order=(2, 0, 0, 12); AIC=2873.301, BIC=2891.325, Fit time=4.597 seconds
Fit ARIMA: order=(2, 1, 1) seasonal_order=(2, 0, 0, 12); AIC=2891.288, BIC=2912.316, Fit time=6.511 seconds
Fit ARIMA: order=(3, 1, 1) seasonal_order=(2, 0, 0, 12); AIC=2881.360, BIC=2905.391, Fit time=8.597 seconds
Fit ARIMA: order=(2, 1, 0) seasonal_order=(1, 0, 0, 12); AIC=2879.303, BIC=2894.322, Fit time=0.579 seconds
Fit ARIMA: order=(2, 1, 0) seasonal_order=(2, 0, 1, 12); AIC=2875.058, BIC=2896.086, Fit time=8.207 seconds
Fit ARIMA: order=(3, 1, 0) seasonal_order=(2, 0, 0, 12); AIC=2866.267, BIC=2887.294, Fit time=8.301 seconds
Fit ARIMA: order=(4, 1, 1) seasonal_order=(2, 0, 0, 12); AIC=2870.532, BIC=2897.567, Fit time=10.208 seconds
Fit ARIMA: order=(3, 1, 0) seasonal_order=(1, 0, 0, 12); AIC=2870.467, BIC=2888.491, Fit time=0.789 seconds
Fit ARIMA: order=(3, 1, 0) seasonal_order=(2, 0, 1, 12); AIC=2867.919, BIC=2891.951, Fit time=7.218 seconds
Fit ARIMA: order=(4, 1, 0) seasonal_order=(2, 0, 0, 12); AIC=2858.880, BIC=2882.912, Fit time=10.408 seconds
Fit ARIMA: order=(5, 1, 1) seasonal_order=(2, 0, 0, 12); AIC=2868.950, BIC=2898.990, Fit time=20.080 seconds
Fit ARIMA: order=(4, 1, 0) seasonal_order=(1, 0, 0, 12); AIC=2866.318, BIC=2887.345, Fit time=0.806 seconds
Fit ARIMA: order=(4, 1, 0) seasonal_order=(2, 0, 1, 12); AIC=2863.844, BIC=2890.879, Fit time=25.681 seconds
Fit ARIMA: order=(5, 1, 0) seasonal_order=(2, 0, 0, 12); AIC=2860.023, BIC=2887.059, Fit time=14.314 seconds
Total fit time: 138.326 seconds

print(__doc__)

# Author: Taylor Smith <taylor.smith@alkaline-ml.com>

import pmdarima as pm
import numpy as np
from matplotlib import pyplot as plt

# #############################################################################
# Load the data and split it into separate pieces
data = pm.datasets.load_wineind()
train, test = data[:150], data[150:]

# Fit a simple auto_arima model
arima = pm.auto_arima(train, error_action='ignore', trace=1,
                      seasonal=True, m=12)

# #############################################################################
# Plot actual test vs. forecasts:
x = np.arange(test.shape[0])
plt.scatter(x, test, marker='x')
plt.plot(x, arima.predict(n_periods=test.shape[0]))
plt.title('Actual test samples vs. forecasts')
plt.show()

Total running time of the script: ( 2 minutes 18.431 seconds)