Apr-20-2020, 09:08 PM
I'm trying to build out some forecasting model with pmdarima.
When using this grid search it keep defaulting to a SARIMAX model.
auto_arima(df['Column']).summary()
Can someone help me use this grid search to create the best parameters for each type of model? I'm trying to compare the results of each different model. I want to force the specific model.
ARMA
ARIMA
SARIMAX
The suggested model is order=(2,1,3),seasonal_order=(0,0,0,0)). I'm not sure if it really has seasonality with all zeros in the seasonal order.
This code seems to create a ARIMA model but it doesn't seem to be optimal:
stepwise_fit = auto_arima(train['column'],start_p=0,start_q=0,max_p=6,max_q=3,seasonal=False,trace=True)
Any help would be greatly appreciated.
Tim
When using this grid search it keep defaulting to a SARIMAX model.
auto_arima(df['Column']).summary()
Can someone help me use this grid search to create the best parameters for each type of model? I'm trying to compare the results of each different model. I want to force the specific model.
ARMA
ARIMA
SARIMAX
The suggested model is order=(2,1,3),seasonal_order=(0,0,0,0)). I'm not sure if it really has seasonality with all zeros in the seasonal order.
This code seems to create a ARIMA model but it doesn't seem to be optimal:
stepwise_fit = auto_arima(train['column'],start_p=0,start_q=0,max_p=6,max_q=3,seasonal=False,trace=True)
Any help would be greatly appreciated.
Tim