rafat / ctsa

A Univariate Time Series Analysis and ARIMA Modeling Package in ANSI C. Updated with SARIMAX and Auto ARIMA.
https://code.google.com/p/ctsa/
Other
57 stars 25 forks source link

Unappropriate return of SARIMAX model #8

Open leibersp opened 7 months ago

leibersp commented 7 months ago

When using auto ARIMA with no exogenous variables (r=0, xreg=NULL), in the case of differencing (d+D) equals 1 and a constant term being allowed, the auto ARIMA will find a model with exogenous variables. It will create helper variables from 1...N. However, if you try to use auto_arima_predict with xreg=NULL it will inevitably fail, as the model will try to access exogenous variables, that are actually not present.

This should not be the case in my opinion, as this disallows using auto arima in a script. (Seg Faults appearing, even though everything is correct).