Many thanks for your quick response and apologies for the misleading
'Subject' line.
Regards
Simon
Friday, September 17, 2004, 11:34:07 AM, you wrote:
NC> Yes, but it is not I think wired into any program.
NC> For example, you could split your data randomly
NC> gen fit = uniform() > 0.5
NC> and then fit the model to one subset
NC> nl ... if fit
NC> and then look at predictions for the other:
NC> ... if !fit
NC> Nick
NC> [email protected]
NC> Simon Moore
>> I am hoping to use non-linear least squares (nl) in Stata v8.0. The
>> analyses are relatively straight forward (nl exp2 var1 var2). However
>> it would be nice if I could perform some form of split half analysis
>> (i.e. fit the model to one half of the dataset and then see how it
>> predicts on the other half). Is there a means of achieving this in
>> Stata?
NC> *
NC> * For searches and help try:
NC> * http://www.stata.com/support/faqs/res/findit.html
NC> * http://www.stata.com/support/statalist/faq
NC> * http://www.ats.ucla.edu/stat/stata/
--
Best regards,
Simon mailto:[email protected]
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
