Not what you might think.
Take the MNL model (with IIA etc) and add a probit link. It is odd that they would call that multinomial probit. Most people would understand “multinomial probit” as a model for multinomial outcomes with multivariate normal disturbances (in general, with non-zero covariances). A screen full into the documentation for Stata’s mprobit it says:
The error terms are assumed to be independent, standard normal, random variables. See [R] asmprobit for the case where the latent-variable errors are correlated or heteroskedastic and you have alternative-specific variables.
A trap for the hasty, the unwary…?
This said, the “real” multinomial probit function in Stata uses some funky options for controlling he nasty integrations necessary to evaluate the likelihood for this model (i.e., simulated MLE a la GHK, with an option for Halton sequences to drive the quasi-Monte Carlo integration). It would be fun to play with this Stata function, alongside the fully Bayesian/MCMC implementations in R for this model (e.g., MNP and bayesm).
A recent e-mail correspondent writes:
I have a degree in applied statistics, and I’m really interested in the lectures notes you put on your website about Bayesian approaches and simulations. That’s something i need to discover and it looks really rich and interesting. I also use R on a very regular basis.
The purpose of this email is that I’m using LaTeX to write some documents, and i can’t find anything on how to install the FF Meta police, which is very clear and easy to read.
Have you anything about that by any chance?
First of all, I should take the Bayes notes down and point you in the direction of The Book (done!).
On FFMeta, I don’t quite get the references to “FF Meta police”. But here is how I did it (below the fold).
Some slides from a talk I gave at a conference sponsored by the American Political Science Association on “Democracy Audits and Governmental Indicators” at the University of California, Berkeley, October 30-31, 2009. The graphic below shows the estimates of country-level democracy for the year 2000 (with marginal 95% credible intervals) that Shawn Treier and I estimated using the Polity IV indicators (a better quality version appears in the slides).
Plus an early attempt at cross-national measurement of regime type (complete with uncertainty bounds), a nice parting gift from our host, Henry Brady.
Into Bayes? Into music? Into wicked coding? Want to live in France for a couple of years? Read on below the fold. (more…)
You know the job has finished running when the fans on the computer get quieter…
Stanford Stats Seminar today is on Lessons from the Netflix Prize, by Robert Bell from AT&T Labs.
1. grep(pattern,x) works when x is a list, with each component being a vector of character strings. The returned argument is a vector of indices, the components of x in which the pattern is matched.
2. in library(foreign), read.spss has an option
trim.factor.names: Logical: trim trailing spaces from factor levels?
Wish I’d RTFMd this earlier, in that "Democrat" != "Democrat "
I installed 10.6 Snow Leopard on my office machine. Seemed to go ok. I then installed 10.6 Server Admin Tools, but have run into the issue reported here. Hmmm.
I wonder if my install of the OS upgrade didn’t go correctly, and that some other discoveries lurk.
People who are regularly bombarded with several streams of electronic information do not pay attention, control their memory or switch from one job to another as well as those who prefer to complete one task at a time, a group of Stanford researchers has found.
I’m editing proofs for my book right now too, and based on an average of about 3 typos per page, I’d have to say I agree with, um, whatever I was reading before the phone/iTunes/vodcast/e-mail/iChat thing popped up.
I’ve made a few graphs summarizing the proposed redistribution in NSW. As I did with the Qld proposed redistribution, I’ve taken Antony Green’s calculations of the notional margins of the proposed divisions, and run some simulations assuming (somewhat implausibly) uniform swing from the 2007 results.
Powered by WordPress
Bad Behavior has blocked 2975 access attempts in the last 7 days.
