{smcl}
{* 28feb2011}{...}
{hline}
help for {hi:grmarg}{right:- v001 28feb2011 jx }
{hline}
{title:Marginal effects and their differences from binary regressions}
{p 8 15 2}{cmd:grmarg} [if] [in] [{cmd:,}
{cmd:x(}{it:variables_and_values}{cmd:)}
{cmd:rest(}{it:stat}{cmd:)}
{cmdab:l:evel(}{it:#}{cmd:)}
{cmdab:del:ta}
{cmdab:boot:strap}
{cmdab:r:eps(}{it:#}{cmd:)}
{cmdab:si:ze(}{it:#}{cmd:)}
{cmd:save}
{cmd:diff}
{cmdab:noba:se}
{cmd:all match dots}
{cmd:dydxmat(}{it:matrix_kxk}{cmd:)}]
{p 4 4 2}
where {it:variables_and_values} is an alternating list of variables
and either numeric values or mean, median, min, max, upper, lower,
previous and {it:stat} is either mean, median, min, max, upper, lower,
previous, grmean(group mean), mrmedian, grmin, grmax.
{title: Description}
{p 4 4 2}
{cmd:grmarg} is a command that returns marginal effects, differences in
marginal effects and their confidence intervals using both {help bootstrap} and {help delta} methods
{title: Options}
{p 4 8 2}
{cmd:x(}{it:variables_and_values}{cmd:)} sets the values of independent
variables for calculating predicted values (marginal effects). The list must alternate variable
names and either numeric values or types of {cmd:stat}.
{p 4 8 2}
{cmd:rest(}{it:stat}{cmd:)} sets the independent variables not specified
in {cmd:x(}{it:variables_and_values}{cmd:)} to one of the types of {cmd:stat}.
{p 4 8 2}
{cmd:level()} sets the {help level} of the confidence interval for predicted
values or probabilities for the commands for which these are provided. The default
is 95.
{p 4 8 2}
{cmdab:del:ta} calculates confidence intervals by the delta method using analytical derivatives.
This method works with cloglog, logistic, logit and probit.
{p 4 8 2}
{cmdab:boot:strap} computes confidence intervals using the bootstrap method. This method takes
roughly 1,000 times longer to compute than other methods. This method works with cloglog, logistic,
logit, and probit.
{p 4 8 2}
{cmd:reps(}{it:#}{cmd:)} specifies the number of bootstrap replications
to be performed. The default is 1000.
{p 4 8 2}
{cmd: size(}{it:#}{cmd:)} specifies the size of the samples to be drawn.
The default is e(N), the same size as the estimation sample.
{p 4 8 2}
{cmd:save} saves current values of independent variables and predictions
for computing changes using the diff option.
{p 4 8 2}
{cmd:diff} computes difference between current predictions and those that
were saved.
{p 4 8 2}
{cmd:nobase} suppresses inclusion of the base values of x in the output.
{p 4 8 2}
{cmd:all} specifies that any calculation of means, medians, etc., should
use the entire sample instead of the sample used to estimate the model.
{p 4 8 2}
{cmd:match} requests {cmd:grmarg} to resample from each category group
of the dependent variable in proportion of the resample size to the original
sample size.
{p 4 8 2}
{cmd:dots} requests a dot be placed on the screen at the beginning of each
replication, thus providing entertainment when a large number of reps() are
requested. It also prints out the percent replications finished.
{p 4 8 2}
{cmd:dydxmat(}{it:matrix_kxk}{cmd:)} supplies a matrix by users of dy/dx or
dxb/dx, where there is polynomial or interaction terms.
{title:Examples}
{p 4 4 2}
To compute the predicted marginal effects, differences in marginal effects, and their confidence intervals using the delta method
for a logit model. All independent variables are held at their means except for black and education specifed in x().
{p 4 8 2}{cmd:.logit vote black educ income}
{p 4 8 2}{cmd:.grmarg, x(black=1 educ=16) save}
{p 4 8 2}{cmd:.grmarg, x(black=0 educ=16) diff}
{p 4 8 2}
:::
{hline}
{p 2 4 2}Authors: Jun Xu{p_end}