Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

AW: st: Creating choice indicator!


From   Peter Grand <grand@ihs.ac.at>
To   "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
Subject   AW: st: Creating choice indicator!
Date   Tue, 20 Mar 2012 18:43:18 +0000

Dear Jean-Francois!

This is exactly my intention, and already showed me that my evaluator is poorly written (although I do not know why by now) and it is no problem with my data.

Again, thank you all! I have also learned a lot in terms of creating different random variables.

Best,

Peter Grand
grand(at)ihs.ac.at
Institute for Advanced Studies
Stumpergasse 56
1060 Vienna

________________________________________
Von: owner-statalist@hsphsun2.harvard.edu [owner-statalist@hsphsun2.harvard.edu]&quot; im Auftrag von &quot;Jean-Francois.Bertrand@fin.gc.ca [Jean-Francois.Bertrand@fin.gc.ca]
Gesendet: Dienstag, 20. März 2012 19:37
Bis: statalist@hsphsun2.harvard.edu
Betreff: RE: st: Creating choice indicator!

Hello,

If I understand correctly, this will help Peter testing his choice model.
Expanding the data provides an environment similar to panel type data.
When programming his function in -ml-, only one observation "per panel" will contribute to the maximum likelihood.

In my experience, the long data structure or shape or form (when possible)
is much faster when programming choice model in -ml-.

Why?

Cheers,
Jean-François Bertrand


-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Nick Cox
Sent: March 20, 2012 2:24 PM
To: 'statalist@hsphsun2.harvard.edu'
Subject: RE: st: Creating choice indicator!

Alternatively, look upwards!

ceil(5 * runiform())

Nick
n.j.cox@durham.ac.uk

Scott Merryman

For a random integer between 1 and 5 the code should be :

 1 + floor(5*runiform())

On Tue, Mar 20, 2012 at 12:32 PM, Peter Grand <grand@ihs.ac.at> wrote:
> @ Scott: Exactly, thank you very much!
>
> @ Joerg: Sorry, if I have not been able to correctly specify my request.
>
> Again, thank you for your efforts!
>
Scott Merryman [scott.merryman@gmail.com]

> Something like this?
>
>
> clear
> set obs 5
> gen id = _n
> expand 5
> sort id
>
> bys id: gen order = _n
> gen chosen = 0
> levelsof id, local(levels)
> foreach l of local levels {
>        if runiform() < .5 {  //probability that id is chosen
>        local a = round(1 +4*runiform(),1)
>        qui replace chosen= 1 if order ==  `a' & id == `l' //1 of 5
> alternatives is chosen
>
>        }
> }
> drop order
> l


On Tue, Mar 20, 2012 at 11:51 AM, Peter Grand <grand@ihs.ac.at> wrote:

>> How can I create a random binary variable which indicates either only one chosen alternative or none of them within an id variable, e.g. 5 choices for each individual id :
>>
>>
>> id   chosen
>> 1       0
>> 1       0
>> 1       1
>> 1       0
>> 1       0
>> 2       0
>> 2       0
>> 2       0
>> 2       0
>> 2       0
>> 3       0
>> 3       1
>> 3       0
>> 3       0
>> 3       0
>>

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index