1

I am trying to apply t a SARIMAX model to predict monthly sales, but when I try to fit the model i have this warning:

Too few observations to estimate starting parameters for seasonal ARMA, All parameters except for variances will be set to zeros.

Even when the dataset shows a clearly seasonality.

Seasonal_Decompose: enter image description here

I've used a Stepwise search to find the best model orders but still having the warning and pretty bad RMSE compared to the test data.

stepwise_model = auto_arima(df_arima['sales_diff'],
                           start_p=1, start_q=1,
                           max_p=3, max_q=3, m=12,
                           start_P=0, seasonal=True,
                           d=1, D=1, trace=True,
                           error_action='ignore',  
                           suppress_warnings=True, 
                           stepwise=True)

PD, the original data is non-stationary so have to work with the differencing to make it stationary.

Any tip to work around that?

Luis Jacobo
  • 97
  • 1
  • 14
  • 2
    I think the default values for seasonal models require 5 full seasons. When the time series is shorter than this, it uses a very naive guess of 0 for parameters. This doesn't mean that the model is not correctly estimated though. It is just a warning, and you should be a bit careful to ensure that you are at the MLE. – Kevin S Mar 22 '21 at 16:59

0 Answers0