Dear Sven,

The feasible GLS comes from Wooldridge's Introductory Econometrics: A Modern Approach, 4th ed., Chapter 8, example 8.7. Actually, Wooldridge two ways of correcting for heteroskedasticity.
The fgls.gfn package was intended to implement those two ways (but the other is not finished yet).

I wrote fgls.gfn because I can't use gretl's default "wls" or "hsk" to replicate the example in Wooldridge's example. That is why.

I did not attached Wooldridge's smoke.gdt since I am not sure the data is proprietary or not. Though the data was obtained from gretl official site (, wooldridge_data.exe.

Will fgls.gfn be back on gretl's server? Thanks.


2015-07-13 15:13 GMT+08:00 Sven Schreiber <>:
Hi Yi-Nung,

can you be more specific? (Book title, chapter, page etc.)

In any case, the example script fails. I understand that you assume that
the Wooldridge data is installed, but that is not ok for our function
package policy. You can add an example data file in a function package
nowadays (in zip format / adding "resources").