Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


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

st: Re: Margins + Marginsplot + xtreg (Not Estimable)


From   John Litfiba <cariboupad@gmx.fr>
To   statalist@hsphsun2.harvard.edu
Subject   st: Re: Margins + Marginsplot + xtreg (Not Estimable)
Date   Tue, 24 Jan 2012 16:16:33 +0100

Well of course
the interaction terms writes i.VAR#i.BIN

and you should read the command as :
> margins, dydx(VAR) at(BIN=(1 2 3 4)) vsquish

;-)

On 24 January 2012 16:06, John Litfiba <cariboupad@gmx.fr> wrote:
> Dear Statalist,
>
> I am struggling with the margins stata function.
>
> I run a fixed effect estimation on a panel dataset using xtreg
> clustered by individuals in my sample.
>
> In the equation I specify an interaction term as : VAR#BIN
> where VAR is a dummy variable that indicates treatment or not
> and BIN is a categorical variable (4 values : 1 2 3 4) that cut the
> variable Volatility in 4 non overlapping bins according to the
> percentiles: for example BIN=2 corresponds to values of volatility
> between the 25th and the 50th percentile, BIN=3 for vol between the
> 50th and 75th percentile, etc
>
> I would like to graph this categorical by categorical interaction
> (with confidence intervals) with marginsplot, but the margins commands
> doesnt work
>
> margins, dydx(trader) at(vol_cat=(1 2 3 4)) vsquish
>
> return not estimable for each value...
>
> However when I do not use bins, but only the interaction
> VAR*c.volatility everything works fine...
>
> What is wrong there ?
> Many thanks
*
*   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 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index