Dear All,
I am currently running an ARMAX model in STATA with an ARMA disturbance process. I have employed different models for the disturbance, and upon comparing the output from STATA with that from SAS, I find that the estimates are nearly identical in most cases. However, for a specific model, I have observed differences in the seasonal MA term values between STATA and SAS. Specifically, in absolute terms, STATA reports 0.3621697 while SAS reports 0.20275 for one MA seasonal term, and STATA reports 0.780338 while SAS reports 0.66553 for another. All other estimates are consistent across both platforms.
I am uncertain whether STATA might offer more reliable estimates in this case, or if there could potentially be an issue with my code or the accuracy of convergence that needs improvement. I have attached my dataset as an Excel spreadsheet, and the STATA code I am using is as follows:
Thank you very much.
I am currently running an ARMAX model in STATA with an ARMA disturbance process. I have employed different models for the disturbance, and upon comparing the output from STATA with that from SAS, I find that the estimates are nearly identical in most cases. However, for a specific model, I have observed differences in the seasonal MA term values between STATA and SAS. Specifically, in absolute terms, STATA reports 0.3621697 while SAS reports 0.20275 for one MA seasonal term, and STATA reports 0.780338 while SAS reports 0.66553 for another. All other estimates are consistent across both platforms.
I am uncertain whether STATA might offer more reliable estimates in this case, or if there could potentially be an issue with my code or the accuracy of convergence that needs improvement. I have attached my dataset as an Excel spreadsheet, and the STATA code I am using is as follows:
Code:
gen Date = _n tsset Date arima Yt X1t X2t, arima(1,1,1) sarima(1,1,2,12) nocons, if Date<=168
