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 <[email protected]>
To
[email protected]
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 <[email protected]> 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/