R help - Aug 2007 - Data simulation with R

Hi, 

I am currently on a placement here at GSK for my studies, and I'm working on
Heckman Models. I have to make simulations, in order to see whether these
models are efficient or not.

 I have to generate a dataset under the following constraints: 

- outcome is 0 or 1 
- one control group, one treatment group, there must be no treatment effect 
- generate one continuous variable and one binary variable which are both
perfectly correlated to treatment, but not correlated to outcome 
- generate one continuous variable and one binary variable which are both
perfectly correlated to outcome, but not correlated to treatment 
- generate one continuous variable and one binary variable which are both
perfectly correlated to both treatment and outcome 

Problem is, it's my first time generating data, so I'm having quite a
hard
time... I usually use SAS, but I didn't manage to generate data using it.
Some people told me it's quite easy to do with R, but I'm not really
used to
this program. Is it really possible to simulate this kind of data with R ?
If it is, can someone help me ?

Thanks in advance. 
-- 
View this message in context:
http://www.nabble.com/Data-simulation-with-R-tf4353531.html#a12405063
Sent from the R help mailing list archive at Nabble.com.

Hi Matiou:

On Thursday 30 August 2007 13:25, Matiou wrote:
> Hi,
>
> I am currently on a placement here at GSK for my studies, and I'm
working
> on Heckman Models. I have to make simulations, in order to see whether
> these models are efficient or not.
>
>  I have to generate a dataset under the following constraints:
>
> - outcome is 0 or 1
> - one control group, one treatment group, there must be no treatment effect
> - generate one continuous variable and one binary variable which are both
> perfectly correlated to treatment, but not correlated to outcome
> - generate one continuous variable and one binary variable which are both
> perfectly correlated to outcome, but not correlated to treatment
> - generate one continuous variable and one binary variable which are both
> perfectly correlated to both treatment and outcome
>
> Problem is, it's my first time generating data, so I'm having quite
a hard
> time... I usually use SAS, but I didn't manage to generate data using
it.
> Some people told me it's quite easy to do with R, but I'm not
really used
> to this program. Is it really possible to simulate this kind of data with R
> ? If it is, can someone help me ?
For an example, please take a look at the "example" section of 
function "selection" in the "micEcon" package.

HTH,
Arne
> Thanks in advance.
-- 
Arne Henningsen
Department of Agricultural Economics
University of Kiel
Olshausenstr. 40
D-24098 Kiel (Germany)
Tel: +49-431-880 4445
Fax: +49-431-880 1397
ahenningsen at agric-econ.uni-kiel.de
http://www.uni-kiel.de/agrarpol/ahenningsen/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url :
https://stat.ethz.ch/pipermail/r-help/attachments/20071004/fd96ff44/attachment.bin