{smcl} {* 11aug2005}{...} {hline} help for {hi:mehetprob} {right: (version 1.0.0) } {hline} {title:Compute marginal effects at means an their standard errors after {help hetprob}} {p 8 20}{cmd:mehetprob} [, {cmdab:nod:iscrete}] {title:Description} {p}{cmd:mehetprob} computes marginal effects at means and their standard errors in a heteroscedastic probit model. {cmd:mehetprob} is executed after Stata's {help hetprob} command. It uses the coefficients and standard errors returned by {help hetprob} in order to compute marginal effects at means and their standard errors by using the delta method. The syntax is such that it requires at least Stata 8. {p}{cmd:mehetprob} delivers the same results as {p 4 8}{inp:. mfx compute} after {help hetprob}. {p}While {help mfx} computes the derivatives needed to infer the standard errors numerically, {cmd:mehetprob} computes them analytically and is therefore considerably faster. It takes only a few seconds while {cmd:mfx compute} after {help hetprob} can take several hours in Stata 8 and several minutes in Stata 9. {title:Options} {p 0 4}{cmdab:nod:iscrete} specifies that dummy (indicator) variables be treated as continuous. {title:Saved results} {col 8}{cmd:XBAR}{col 20}The matrix {cmd:XBAR} contains means of the regressors of the main equation of the {col 20}heteroscedastic probit model. {col 8}{cmd:ZBAR}{col 20}The matrix {cmd:ZBAR} contains means of the regressors of the equation to model the {col 20}variance in the heteroscedastic probit model. {col 8}{cmd:DER}{col 20}The element (i,j) of the matrix {cmd:DER} contains the derivative of the marginal effect of the {col 20}regressor w_i with respect to the regression coefficient theta_j. {col 20}The delta method is based on this matrix. {col 8}{cmd:C}{col 20}The matrix {cmd:C} contains the results: marginal effects, standard errors, {col 20}z-values and P-values derived therefrom. {col 8}{cmd:mehetprob} temporarily creates the matrices C, C1, C2, ZG, ZGx, XB, XBx, B, ZBAR, XBAR, G, D, V. {col 8}If any of this matrices exists prior to the execution of the command, it will be overwritten. {title:Example} {p 4 8}{inp:. hetprob x y, het(y z)} {p 4 8}{inp:. mehetprob} {title:Author} Thomas Cornelißen, University of Hannover, Germany cornelissen@mbox.iqw.uni-hannover.de {title:Also see} {p 0 9}On line: help for {help hetprob}, {help mfx}.