Re: st: interactions, slopes, and standard errors

["Jesper B. Sorensen" <sorensen@mit.edu>]

Here's a stab at a hack that would do it for the 25th, 50th and 75th 
percentiles.  For arbitrary values you could change the forvalues to a 
foreach and simply list the values you want (and change the display'ed 
text) or have this be something the user supplies.


program marginalx2, rclass
version 8.1
syntax , x(varname) y(varname) xy(varname)

qui sum `y' if e(sample), det

forvalues p = 25(25)75 {

        local marginal = _b[`x'] + _b[`xy']*r(p`p')

        matrix A = e(V)
        matrix B1 = A["`x'" , "`x'"]
        local var_b1 = B1[1,1]

        matrix B2 = A["`xy'" , "`xy'"]
        local var_b2 = B2[1,1]

        matrix B3 = A["`x'" , "`xy'"]
        local covar_b1b2 = B3[1,1]

        local variance = `var_b1' + (r(p`p')^2)*`var_b2' + 
2*r(p`p')*`covar_b1b2'

        dis ""
        disp in gr "Marginal effect of `x' at the " `p' "th percentile of 
`y' = " `marginal'
        disp in gr "Variance of marginal effect of `x' at the " `p' "th 
percentile `y' = " `variance'

        return scalar margx`p' = `marginal'
        return scalar var_margx`p' = `variance'
}
end








At 12:13 PM 3/26/2004 -0600, you wrote:

> Has anyone written a general program to calculate and display the 
> estimated slope and associated error of Y on x1 at various values of x2, 
> where Y = b0 + b1x1 + b2x2 + b3x1x2
>
> Last fall Scott Merryman posted a nice program (which I copy in, below) 
> which takes care of this at the mean of x2. Thanks, Scott!, We could also 
> be interested in doing it for other values.
>
> Chip Hunter
>
> **
> - Scott Merryman's program to calculate the marginal effects and the 
> variance of interaction terms -
>
> program marginalx, rclass
> version 8.1
> syntax , x(varname) y(varname) xy(varname)
>
> sum `y' if e(sample), meanonly
>
> local marginal = _b[`x'] + _b[`xy']*r(mean)
>
> matrix A = e(V)
> matrix B1 = A["`x'" , "`x'"]
> local var_b1 = B1[1,1]
>
> matrix B2 = A["`xy'" , "`xy'"]
> local var_b2 = B2[1,1]
>
> matrix B3 = A["`x'" , "`xy'"]
> local covar_b1b2 = B3[1,1]
>
> local variance = `var_b1' + (r(mean)^2)*`var_b2' + 2*r(mean)*`covar_b1b2'
>
> dis ""
> disp in gr "Marginal effect of `x' at the mean of `y' = " `marginal'
> disp in gr "Variance of marginal effect of `x' at the mean of `y' = " 
> `variance'
>
> return scalar margx = `marginal'
> return scalar var_margx = `variance'
> end
>
>
>
>
> **************************************************
>
> Chip Hunter
> University of Wisconsin-Madison
> School of Business
> Management and Human Resources Dept.
> 4114 Grainger Hall
> 975 University Ave.
> Madison, WI 53706
> 608-265-3533
> **************************************************
>
> *
> *   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/
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Jesper B. S�rensen
Richard S. Leghorn (1939) Associate Professor of Strategic Management
Sloan School of Management
Massachusetts Institute of Technology
E52-581
Cambridge, MA 02142
http://web.mit.edu/sorensen/www/
(617) 253 7945  -- voice
(617) 253 2660  -- fax 

*
*   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/