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 on April 23, and its replacement, statalist.org is already up and running.


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

Re: st: on-the-fly indicator at cutpoints of continuous variable


From   Nick Cox <njcoxstata@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: on-the-fly indicator at cutpoints of continuous variable
Date   Tue, 20 Nov 2012 18:37:56 +0000

In your original example it is unclear whether k is a constant you
give or it's a parameter to be estimated. But either way -nl- can do
it.

I suppose we have all wanted to be able to write e.g. transformations
on the command line. My wild guess is that the difficulty is not so
much writing code to accept such input, but working out what that
would imply all the way through to r- and e-class results, -margins-,
-marginsplot-, etc., etc.

I have no idea whether it's on the development list for Stata 13 or
any other version. But I suspect that there are many things that users
really, really miss that are more crucial. To each their own
preferences, naturally.

Nick

On Tue, Nov 20, 2012 at 4:50 PM, Hoffman, George <ghoffman@mcw.edu> wrote:
> Ah yes, nl. I have done that. I need to write a wrapper to make it easier. And that linear regression was just an example; a more generic approach would maybe go on the Stata 13 wish list!
> Thanks again,
> George
>
>
> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Nick Cox
> Sent: Tuesday, November 20, 2012 9:57 AM
> To: statalist@hsphsun2.harvard.edu
> Subject: Re: st: on-the-fly indicator at cutpoints of continuous variable
>
> You can do something close to this in -nl-, but you would need to do it differently.
>
> Nick
>
> On Tue, Nov 20, 2012 at 3:44 PM, Hoffman, George <ghoffman@mcw.edu> wrote:
>
>> I'd like to model the following:
>>                 reg Y  c.x1 c.x2 i.x3
>> where i.x3 takes on a value of 1 if x3<k, otherwise 0
>>
>> this does not work:
>>                 reg Y c.x1 c.x2 i.(x3<k)
>>
>> it is simple to create an indicator variable and include that in the regression, but is there a way to include an on-the-fly indicator in commands?
>>
>> Apparently this is 'easy' in R. is there an equivalent in Stata?
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


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