Command for including trend in seemingly unrelated regression in stata.
I understand the question, but I would need some more information to answer this question.
Even I may be wrong if at all we can include trends in seemingly unrelated regression.
Giving it a try if this helps
1. The different coefficients from the ones from the original equations can be used. (If this is what you are looking for)
2. I have heard about Haussman test that may be used in this case. The difference in Haussman test is test considers as true the first equation and not the second equation as normal ones consider
If the answer is not related to what you have expected, apologies for the same :)
Honestly I thought that trend was a function of the data and I wonder how it could be expressed as an IV in a regression. Perhaps you can explain further. Best wishes, David Booth
Thank you very much for the willingness to help. Actually, the trend is not part of my variables but since its time series analysis, I wish to include the trend just like we are able to include trend when we do ADF or PP unit root tests and even ARDL regressions especially using Eviews. But for this particular case, I don't just know the command in Stata to include the trend. I hope I have brought clarity enough to my question. David Eugene Booth
Thank you very much. I am actually using two variables, electricity consumption and GDP growth. So I have two equations with each variable as a dependent variable and regressor at the same time. But I wish to include a trend to both equations in the Stata estimation but I am unable. I don't know if this is an added information enough to guide you. Ram Prakash
Dear Adamu Braimah Abille, in Stata, you can generate trends using filters into the original series (e.g. Hodrick and Prescott). I recommend the command
https://www.stata.com/manuals/tstsfilter.pdf
I hope it helps,
- Badges
- Science topic
- Similar topics
- Statistics
- Time Series