turn on suggestions

Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

Showing results for

Find a Community

- Home
- /
- Analytics
- /
- Stat Procs
- /
- How to do feasible generalized least square in pro...

Topic Options

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

06-10-2016 08:24 PM - edited 06-10-2016 08:30 PM

Hi all,

I was trying to do a feasible generalized least square (FGLS) in SAS 9.2 to adjust for heteroscedasticity. The form of the variance is unknown. I am just trying to use the residuals to estimate the variance and then do an iterative reweighting (I think this is what FGLS is).

I googled it and found the SAS documentation here. But the documentation didn't explain very clearly how to implement FGLS in proc model. It only says using H.var. I tried the following code,

```
proc model data=temp ;
y=b0+b1*x1+b2*x2;
h.y=resid.y**2;
fit y /itprint ;
run;
```

but SAS says:

*WARNING: Can only do FIML or GMM estimation when parameters are shared by the mean model and the variance model. The estimation requested will ignore the variance model.*

So I assume the code is not doing FGLS. Can anyone provide an example code that will do a FGLS? I am OK with either proc model or any other procedures that will do the job.

Thank you all so much.

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

06-13-2016 06:25 PM

You might be better putting this post in the Forecasting and Econometrics community since PROC MODEL is part of SAS/ETS. I am not too familiar with PROC MODEL for this application, but your syntax does not agree with the examples in the documentation you cite. Because your resid.y variable involves the b0, b1, b2 parameters, by definition (since y^ can only be determined if one has the parameters), your variance function shares parameters with the model for y. The documentation says that you can't do that for FGLS.

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

06-14-2016 12:03 PM

Thanks, I will repost it