{smcl} {* 06sep2006/08aug2006}{...} {hline} help for {hi:ddirifit} {hline} {title:Displaying discrete changes and marginal effects from dirifit output} {p 8 17 2} {cmd:ddirifit} [{cmd:,} {cmd:at(}{it:variables_and_values}{cmd:)} {cmdab:iter:ate(}{it:#}{cmd:)} ] {title:Description} {p 4 4 2} {cmd:ddirifit} displays the change in predicted dependent variable for three types of discrete changes in explanatory variable and the marginal effects: {p 8 8 2}{cmd:Discrete Changes} {p 8 8 2}{cmd:Min --> Max} shows the change in predicted dependent variable when the explanatory variable changes from its minimum value to its maximum value, while keeping all other explanatory variables at their specified values (by default the mean). This is the only effect shown for dummy variables (variables with only two distinct values). {p 8 8 2}{cmd:+-SD/2} shows the change in predicted dependent variable when the explanatory variable moves from half a standard deviation below its specified value (by default the mean) to half a standard deviation above its specified value, while keeping all other variables at their specified values. In other words: It shows the effect of a standard deviation change in explanatory variable, centered around the specified value, on the predicted dependent variable. {p 8 8 2}{cmd:+-1/2} shows the change in predicted dependent variable when the explanatory variable moves from half a unit below its specified value to half a unit above its specified value, while keeping all other variables at their specified values. In other words: It shows the effect of a unit change in explanatory variable, centered around the specified value, on the predicted dependent variable. {p 8 8 2}{cmd:Marginal Effects} {p 8 8 2}{cmd:MFX at x} shows the marginal effect of each (non dummy) variable, while keeping all variables at their specified values. The marginal effect is the change in predicted dependent variable for a unit change in the explanatory variable, assuming that the effect doesn't change over that interval. {p 4 4 2} {cmd:ddirifit} is only allowed after {cmd:dirifit} in the alternative parameterization, i.e. if one or more of {cmd:muvar()}, {cmd:mu1|2|3|...|}{it:k}{cmd:()}, {cmd:baseoutcome()}, and {cmd:phivar()} is specified or if the {cmd:alternative} option is specified. {title:Options} {p 4 8 2}{cmd:at(}{it:variables_and_values}{cmd:)} allow the user to specify at which values of the explanatory variables the effects are calculated. {it:variables_and_values} is an alternating list of variables and either numeric values or mean, median, min, max, p1, p5, p10 p25, p50, p75, p90, p95, p99. The default is mean for all variables. The statistics p1, p5, ..., p99, are the 1st, 5th, ..., 99th percentile. {p 4 8 2}{cmd:iterate(}{it:#}{cmd:)} is passed directly to {cmd:nlcom}. It specifies the maximum number of iterations used to find the optimal step size in calculating numerical derivatives. You should rarely have to use this option. See: {help nlcom}. {title:Examples} {p 4 8 2}{cmd:. dirifit propfood, mu(income nchild) phi(haspartner)} {p 4 8 2}{cmd:. ddirifit} {p 4 8 2}{cmd:. ddirifit, at(income median nchild 2)} {title:Author} {p 4 4 2}Maarten L. Buis, Vrije Universiteit Amsterdam{break}m.buis@fsw.vu.nl {title:Also see} {p 4 13 2} Online: help for {help dirifit} and {help nlcom}