R for economists (was: [R] Almost Ideal Demand System)

[Ajay Shah]

On Thu, Feb 19, 2004 at 04:40:13PM -0000, Simon Cullen wrote:
> On Thu, 19 Feb 2004 19:06:08 +0530, Ajay Shah <ajayshah at mayin.org> wrote:
> 
> <snip R for economists>
> 
> >Okay, I made a start, with
> >      http://www.mayin.org/ajayshah/KB/R/more.html
> >
> >Tell me what should go into it. :-)
> 
> I think the most important thing for economists would be a list of the  
> common econometric models and how they are accessed in R, for instance  
> packages and the functions within those packages.
> 
> An example is what is known as a Tobit model to econometricians, which is  
> one type of a survival model (to everybody else!). (This isn't that good  
> an example as help.search("tobit") would probably turn up the right  
> library.)
> 
> If you could produce a table with Econometrics to R translations that  
> would be immensely helpful, as a start. I know someone suggested creating  
> a facility for 'aliasing' econometrics-speak to R functions, but this got  
> shot down for being unwieldy (rightly so, I think).

This sounds like a good idea. It'd be neat to have a TeX document
where "many" mainstream econometrics models are written in
mathematics, and then fragments of R code are shown for the purpose.

In addition, I see rolling-your-own-likelihood-function as being a
very common thing in economics today. So I'd just put a lot of focus
on exposition of R facilities for doing MLE, with functionality
comparable with the maxlik and co libraries of gauss. I am not yet
fluent in R, but my early sense is that maxlik+co of gauss are a bit
ahead of what we have in R.

-- 
Ajay Shah                                                   Consultant
ajayshah at mayin.org                      Department of Economic Affairs
http://www.mayin.org/ajayshah           Ministry of Finance, New Delhi