FMOLS

[seires1]

Hi,

I am using the eviews-posted FMOLS program. I think it's awesome, someone posts something like that.
However i am a newby in eviews. I have read the help files and also the already posted question regarding this program.
I think i am doing something wrong:
- First of all i open my eviews workfile, go to open/program and run the program. However the only result i see is that i see
the subroutine fmols etc. in the bottom of the screen.
- Do i first have to call the cov, ses and so on?

Please help me!!!

Greetings,
Johan

[EViews Glenn]

The program only puts the result into the vector for the beta, tstats, and cov that you specify. It doesn't (yet) produce a table with those results.

Note that if you want a full featured FMOLS (and are willing to work with beta software), I highly recommend that you download the beta version for EVIews 7, which has (in my biased opinion) very nice tools for estimating and working with single equation nonstationary regression models.

[seires1]

Could someone explain to me how to activate this program FMOLS ? Do i first have to call the routine on data?
I ve tried to do this by using the command window, but when i type in call fmols(y, g, bw, b, ses, cov) he gives an error...
And if i run the subroutine nothing i happening.
I want to perform the FMOLS for a group of regressor comparing to a dependent variable in a panel structure.
Please help me!

[trubador]

When you open the program in EViews along with your workfile, you should call the subroutine within the program window instead of command line. You have to supply 6 arguments of different types (e.g. series, vectors, etc.) to run the program correctly. There is also an example of code provided in the program that explains how to call the routine on data in an EViews workfile.

[Zeno]

Hi,

I suppose this is the Phillips and Hansen (1990) procedure? I was looking for the Pedroni (2001) FMOLS procedure which is applied to panels. Does anyone know if the existing code is similar to the Pedroni (2001) procedure so that with a little bit of work it could be transformed accordingly?

Thanks

Zeno

References: Pedroni, P. (2001), "Purchasing Power Parity Tests in Cointegrated Panels", Review of Economics and Statistics, 83(4), 727-731.