Datum: Wed, 30 Jul 2008 15:23:18 +0200 Von: "Sven Schreiber" <svetosch(a)gmx.net&gt; An: Gretl development <gretl-devel(a)lists.wfu.edu&gt; Betreff: [Gretl-devel] minor issues

Hello, with 1.7.6rc2 I experience the following minor flaws. The timing of this report is purely coincidental, I'm not a big fan of last-minute fixes without proper testing, so I'm not pushing this to be fixed before 1.7.6. 1) Clicking ok in the dialog window for generating random variables adds the specified variable, but doesn't close the window. Actually I can imagine that this is a feature, not a bug, but just in case it's not... 2) With panel --random-effects estimation the Hausman test of RE vs. FE isn't printed when the test statistic becomes negative. From what I recall of the translated strings, there used to be a warning message instead. I didn't agree with the content of that warning message, but whatever one thinks of those negative statistics, gretl shouldn't just silently ignore them and leave the user wondering why no test output is shown. 3) And finally a related feature request: I think $hausman should not only access the endogeneity tests after tsls, but should also retrieve the Hausman test after --random-effects. thanks, sven _______________________________________________ Gretl-devel mailing list Gretl-devel(a)lists.wfu.edu http://lists.wfu.edu/mailman/listinfo/gretl-devel

Datum: Wed, 30 Jul 2008 17:41:12 +0200 Von: "Sven Schreiber" <svetosch(a)gmx.net&gt; An: Gretl development <gretl-devel(a)lists.wfu.edu&gt; Betreff: Re: [Gretl-devel] minor issues

Oh well, yet another: The following script gives an error: # meta parameters (stacked time series) N = 5000 T = 5 size = N*T nulldata size setobs T 1:1 --stacked-time-series Obviously, setobs doesn't accept a variable for the periodicity. It could be just a missing feature, but in the realm of gretl's versatile scripting language I'm actually inclined to call it a bug. cheers, sven -------- Original-Nachricht -------- > Datum: Wed, 30 Jul 2008 17:16:23 +0200 > Von: "Sven Schreiber" <svetosch(a)gmx.net&gt; > An: Gretl development <gretl-devel(a)lists.wfu.edu&gt; > Betreff: Re: [Gretl-devel] minor issues > One more thing...: > > In a script, the line > > matrix uvec = transp(umat[1,]) > > works, but > > matrix uvec = umat[1,]' > > doesn't. Is this intended? I thought that should be equivalent. > > -sven > > -------- Original-Nachricht -------- > > Datum: Wed, 30 Jul 2008 15:23:18 +0200 > > Von: "Sven Schreiber" <svetosch(a)gmx.net&gt; > > An: Gretl development <gretl-devel(a)lists.wfu.edu&gt; > > Betreff: [Gretl-devel] minor issues > > > Hello, > > with 1.7.6rc2 I experience the following minor flaws. The timing of this > > report is purely coincidental, I'm not a big fan of last-minute fixes > > without proper testing, so I'm not pushing this to be fixed before > 1.7.6. > > > > 1) Clicking ok in the dialog window for generating random variables adds > > the specified variable, but doesn't close the window. Actually I can > imagine > > that this is a feature, not a bug, but just in case it's not... > > > > 2) With panel --random-effects estimation the Hausman test of RE vs. FE > > isn't printed when the test statistic becomes negative. From what I > recall of > > the translated strings, there used to be a warning message instead. I > > didn't agree with the content of that warning message, but whatever one > thinks > > of those negative statistics, gretl shouldn't just silently ignore them > and > > leave the user wondering why no test output is shown. > > > > 3) And finally a related feature request: I think $hausman should not > only > > access the endogeneity tests after tsls, but should also retrieve the > > Hausman test after --random-effects. > > > > thanks, > > sven > > _______________________________________________ > > Gretl-devel mailing list > > Gretl-devel(a)lists.wfu.edu > > http://lists.wfu.edu/mailman/listinfo/gretl-devel > _______________________________________________ > Gretl-devel mailing list > Gretl-devel(a)lists.wfu.edu > http://lists.wfu.edu/mailman/listinfo/gretl-devel _______________________________________________ Gretl-devel mailing list Gretl-devel(a)lists.wfu.edu http://lists.wfu.edu/mailman/listinfo/gretl-devel

Datum: Wed, 30 Jul 2008 12:32:47 -0400 (EDT) Von: Allin Cottrell <cottrell(a)wfu.edu&gt; An: Gretl development <gretl-devel(a)lists.wfu.edu&gt; Betreff: Re: [Gretl-devel] minor issues

On Wed, 30 Jul 2008, Sven Schreiber wrote: > The following script gives an error: > > # meta parameters (stacked time series) > N = 5000 > T = 5 > size = N*T > nulldata size > setobs T 1:1 --stacked-time-series > > Obviously, setobs doesn't accept a variable for the periodicity. > It could be just a missing feature, but in the realm of gretl's > versatile scripting language I'm actually inclined to call it a > bug. The failure here is over-determined: (a) up till now, setobs has not accepted a variable for its first argument, only a numerical value. That's now fixed in CVS.

subsequently, "T" has been deleted by the time setobs is called. We have a "preserve" option for nulldata. At present this only preserves existing matrices. I think it should probably preserve existing scalars too, but I'm not going to attempt that for 1.7.6, I'll put it on the list for 1.7.7.

On Wed, 30 Jul 2008, Sven Schreiber wrote: >> We have a "preserve" option for nulldata. At present this only >> preserves existing matrices. I think it should probably preserve >> existing scalars too, but I'm not going to attempt that for 1.7.6, >> I'll put it on the list for 1.7.7. > Agreed. Do you want me to create a tracker item for this? That's probably a good idea.

This works: matrix uvec = (umat[1,])'

Sure, no problem. It's just that (IIRC) the manual is quite proud of the "natural" ' notation, so some users may expect it to work.

Datum: Wed, 30 Jul 2008 12:36:47 -0400 (EDT) Von: Allin Cottrell <cottrell(a)wfu.edu&gt; An: Gretl development <gretl-devel(a)lists.wfu.edu&gt; Betreff: Re: [Gretl-devel] minor issues

On Wed, 30 Jul 2008, Sven Schreiber wrote: > 1) Clicking ok in the dialog window for generating random > variables adds the specified variable, but doesn't close the > window. Actually I can imagine that this is a feature, not a > bug, but just in case it's not... Intended. That's why there's a "Close" button. > 2) With panel --random-effects estimation the Hausman test of RE > vs. FE isn't printed when the test statistic becomes negative. > From what I recall of the translated strings, there used to be a > warning message instead. I didn't agree with the content of that > warning message, but whatever one thinks of those negative > statistics, gretl shouldn't just silently ignore them and leave > the user wondering why no test output is shown. Fair enough, I'll take a look at that.

> 3) And finally a related feature request: I think $hausman > should not only access the endogeneity tests after tsls, but > should also retrieve the Hausman test after --random-effects. Do you want to add that to the tracker?