EViews.com

Caveats about the asymptotics required for these latter results should be kept in mind. You need the number of observations for each cross-section unit to go to infinity.

It isn't a built-in feature. We have an example program for doing multinomial logit using the logl provided in your EViews example programs directory, but not the multinomial probit, as that is a more difficult specification to estimate...

Gareth is correct as far as one-step features. However what you can do is to use the Generate By Classification proc of a series to create series with the desired binning, then do a crosstab of the classified data. We'll even show the bin definitions in the crosstab since we use the value maps.

The warning message is pretty clear about what is happening. The estimate of your random effects variance is zero...hence the GLS estimates are equal to the OLS estimates. In general, there's nothing really to do here, except perhaps try a different specification or estimator for the random effects ...

I'm a little late to the party, but want to make one other comment on the sample exclusions. An easy way to string together a bunch of NA excludes is to put the variables you want to exclude in a group, say group todrop x y z then set your sample to drop observations that have NAs anywhere in the ro...

We don't have built-in tools, though you could, with a bit of work get it up and running using the logl.

I believe that Stata does offer support.

Before I comment, let me note that we here at QMS reserve the right not to respond to posts on the forum, though we will try to do so when possible, especially when we feel that the comments will be helpful to many. Obviously, we try to monitor the forum and help out, but we simply cannot respond to...

DOLS requires that you include leads and lags of the differenced cointegrating regressors on the right-hand side of your equation (see for example, Hayashi (2000), Econometrics , p. 655 or Zivot and Wang (2006), Modeling Financial Time Series, p. 451. Hamilton (1994) also has a discussion.

That's not entirely correct. You could do the dynamic regression yourself (adding leads and lags as desired) and, as suggested in Zivot and Wang (2006) use Newey-West standard errors to get valid t-stats. This is not generally the variance estimator described in the literature, for example, in, Haya...

There are no in-built tools for this, but in principle you should be able to use the logl object.

Ng and Perron, "A Note on the Selection of Time Series Models,"
Oxford Bulletin of Economics and Statistics, 67, 1 (2005), 115-134.

I should have pointed out that if you really want them, you could go directly to the MacKinnon programs...

http://econ.queensu.ca/faculty/mackinnon/numdist/

I can take a look at this one for you, but before I do, I want to make certain that when you computed the information criteria, you used a common sample for all comparisons...

There isn't, at the moment, a direct way of doing this. The output for the unit root tests will provide the values for the 1%, 5% and 10% values, but we don't have a user-accessible function to return the value. It's on our list of things to look at for future versions -- one issue that we've seen i...

Drop one of your period dummies.

More generally, if you have a lot of periods, use the @expand command in the estimation routine. @expand will allow you to drop one of the time-period dummies automatically.