Re: st: how to express the mean of a variable in a calculation?

[jjc.li@utoronto.ca]

Thank you Joseph and Kit! But there's an error in my calculation:

. di `slpmean1'
.22114205

. nlcom (_b[halflnpllnpl]+slpmean1*slpmean1-slpmean1)/(slpmean1*slpmean1)

slpmean1 not found
r(111);

Hope you can help we with it. Thanks again.

Crystal


Quoting Joseph McDonnell <jockmcdock@gmail.com>:

> quietly summarize x2 if year>=1 & year<=5
> display r(mean)
> local x2mean1_5=r(mean)
> di `x2mean1_5'
>
> you can then use x2mean1_5 in your calculations. You can, in
> principle, use r(mean) but this will be wiped on the next r-class
> command.
>
> If you want the mean for each individual year, you could use a loop
>
> forvalues yr=1(1)5 {
>       quietly summarize x2 if year==`yr'
>        local mean`yr'=r(mean)
> }
>
> di `mean1'
> di `mean5'
>
> Cheers
>
> Joseph
>
> 2009/7/8  <jjc.li@utoronto.ca>:
> > Dear all,
> >
> > I am going to calculate sth after the the regression. I know _b[x1]
> > represents the coef. of variable x1. But how to express the mean of variable
> > x2, in particular just the mean of x2 from year 1 to year 5 (let's say there
> > are total 10 years)?
> >
> > Thanks!
> >
> > Crystal
> >
> > *
> > *   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/