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