Re: [Gretl-devel] RE-Logit using R
On Sat, 8 Nov 2014, Allin Cottrell wrote:
On Sat, 8 Nov 2014, Sven Schreiber wrote:
> given that my first attempts to implement panel logit with so-called
> random effects have been 95% but not 100% successful I have now written a
> wrapper function to outsource this task to R. Given that this is a
> potentially computational-intensive task I think it makes sense to incur
> the overhead of invoking R.
> If somebody likes to test and feed back the results, they're welcome (see
> attached).
[...]
Thanks, Sven. Here's what I'm seeing with 3.1.0:
? bundle myRElogit = RElogitR(binary, rhs)
Loading required package: Matrix
Loading required package: Rcpp
fixed-effect model matrix is rank deficient so dropping 1 column /
coefficient
Error in if (!dims["useSc"] || trivial.y) NA else sigmaML :
missing value where TRUE/FALSE needed
In addition: Warning message:
In storage.mode(dims) <- "integer" : NAs introduced by coercion
External command failed
Ah, but I just updated and it seems to work OK with R 3.1.2:
Random effects:
Groups Name Variance Std.Dev.
groups (Intercept) 0.03334 0.1826
Number of obs: 891, groups: groups, 140
Fixed effects:
Estimate Std. Error z value Pr(>|z|)
(Intercept) 0.721329 0.326635 2.208 0.0272 *
EMP 0.001841 0.004444 0.414 0.6787
WAGE -0.052298 0.013335 -3.922 8.79e-05 ***
Allin