I am trying to predict weekly sales using ARMA ARIMA models. I could not find a function for tuning the order(p,d,q) in statsmodels
. Currently R has a function forecast::auto.arima()
which will tune the (p,d,q) parameters.
How do I go about choosing the right order for my model? Are there any libraries available in python for this purpose?
You can implement a number of approaches:
ARIMAResults
include aic
and bic
. By their definition, (see here and here), these criteria penalize for the number of parameters in the model. So you may use these numbers to compare the models. Also scipy has optimize.brute
which does grid search on the specified parameters space. So a workflow like this should work:
def objfunc(order, exog, endog):
from statsmodels.tsa.arima_model import ARIMA
fit = ARIMA(endog, order, exog).fit()
return fit.aic()
from scipy.optimize import brute
grid = (slice(1, 3, 1), slice(1, 3, 1), slice(1, 3, 1))
brute(objfunc, grid, args=(exog, endog), finish=None)
Make sure you call brute
with finish=None
.
You may obtain pvalues
from ARIMAResults
. So a sort of step-forward algorithm is easy to implement where the degree of the model is increased across the dimension which obtains lowest p-value for the added parameter.
Use ARIMAResults.predict
to cross-validate alternative models. The best approach would be to keep the tail of the time series (say most recent 5% of data) out of sample, and use these points to obtain the test error of the fitted models.