R, ADMB, and SAS

This (and all the follow-ups) is the most interesting live discussion about software comparison recently. I read Zhang et al. paper some time ago but did not find it very interesting. But this ongoing discussion that triggered by that paper is really interesting.

Using externally generated multiply imputed data sets with Zelig

This post provides some hints. This may be what I need to improve my infant mortality paper.

The same can be achieved using the "mi()" function provided by the Zelig package.

Free scientific software

The large number of free scientific software included in the Sage package gives a pretty good idea about what is available.

SabreR

SabreR just released an update. It is another software package that can estimate multivariate multilevel model (other options are aML, MCMCglmm, etc.). They seem to also have a book dedicated to the software, which be worth checking out.

It will be great if the author can incorporate some plotting function into the package.

Mint-coated Debian is great

I really like my Mint-coated Debian system (running xfce). For experimentation, one can simply enable the Debian testing software repository; for production use on my primary workstation, it may be a good idea to use the Mint software repository because the Mint folks promised to ensure that the update pack they issued won't break the system.

Sakura

On my Mint Xfce box, the xfce terminal started and responded more slowly than gnome terminal. This is quite strange and I have been using the gnome terminal instead. Today I found sakura, another gtk-based terminal emulator and really like it.

Upgrading Mint Debian with Debian source

I finally got tired waiting for the Mint guy to release their next update pack. I just added the Debian testing source and did a complete update. The system is running well so far and shows no sign of breakage.

The new StatET looks promising

StatET plugin for Eclipse just reached its 2.0 release, which is a big jump from the previous one. The new version seems to have quite a few new features that worth being considered seriously.

I like Emace+ESS, Rstudio, and Eclipse+StatET. Choice is a good thing!

Interesting "emacs vs. vi" blog

This one actually has some quite useful information ...

A few free books

1. Advanced Linux Programming;
2. Free for All;
3. Two Bits;
4. ...

Good R technical resources

Can be found here.

Doing email from within emacs

Here is a tutorial. A different one here.

Performance difference between Stata and R

With respect to multinomial logit model, the performance difference between the two packages are quite large, based on this post.

Stallman's comment on Jobs's death

Here. Very short and controversial. Here is a good analysis. Comments from Eric Raymond.

This guy, Larry, tried to start a war but got beaten up pretty bad.

Martin jetpack

Isn't this cool?

GEE using Stata vs. R

I am running GEE logistic regression model for my fetal loss paper. As usual, I compare results between Stata and R and make sure they are consistent. To my surprise, the models assuming independent correlation structure give similar results but the models assuming exchangeable correlation structure give drastically different results.

It turns out that there is only one woman in my sample who reported a total number of eleven pregnancies (all others reported ten or less) and the presence of this single observation had huge influence on the algorithm used in R but not the one used in Stata. After excluding this single observation, the two sets of results look identical.

Maximum likelihood example

Here is one and here is another one. This one is also helpful.

Updated examples that combine Rcpp and CppBugs

This example is very informative in illustrating how to use CppBugs module with Rcpp to conduct fast MCMC simulation in R.