{smcl}
{hline}
help for {cmd:dfl}{right:Joao Pedro Azevedo}
{hline}
{title:{cmd:DiNardo, Fortin and Lemieux Counterfacual Kernel Density}}
{p 8 12 2}{cmdab:dfl}
{it:depvar} {it:indepvars}
[{cmdab:if}{it: exp}]
[{cmdab:in}{it: range}]
{cmd:,}
{cmdab:outcome}{cmd:(}{it:varname}{cmd:)}
[{cmdab:group}{cmd:(}{it:varname}{cmd:)}
{cmdab:min}{cmd:(}{it:integer}{cmd:)}
{cmdab:max}{cmd:(}{it:integer}{cmd:)}
{cmdab:nbins}{cmd:(}{it:integer}{cmd:)}
{cmdab:w}{cmd:(}{it:bandwidth}{cmd:)}
{cmdab:step}{cmd:(}{it:varlist}{cmd:)}
{cmd:adaptive}
{cmd:gauss}
{cmd:epan}
{cmd:oaxaca}
{cmd:quietly}
{cmd:probit}
{cmdab:nxvar(}{it:string}{cmd:)}
{cmdab:ncfactual(}{it:string}{cmd:)}
{cmdab:nufactual(}{it:string}{cmd:)}
{cmdab:nfactual(}{it:string}{cmd:)}
{it:{help graph_combine}}
{it:{help axis_selection_options}}
{it:{help axis_scale_options}}
{it:{help title_options}}
{cmd:nogroupt}
{cmdab:name3(}{it:string}{cmd:)}]
{title:Description}
{p 4 4 2}{cmd:dfl} estimates DiNardo, Fortin and Lemieux Counterfacual Kernel Densities.
Treatment status is identified by depvar==1 for the treated and depvar==0
for the untreated observations.{p_end}
{p 4 4 2}The propensity score - the conditional treatment probability - is estimated by
the program on the indepvars.{p_end}
{p 4 4 2}{cmdab:outcome} variable has to be specified.{p_end}
{p 4 4 2}{cmd:dfl} also support graph options such as {it:{help graph_combine}}, {it:{help title_options}} and
{it:{help axis_selection_options}}.{p_end}
{p 4 4 2}Several different weight can be used on this procedure. In this ado we use the following weight.
Weight=1-Prob(Depvar=1)/Prob(Depvar=0).{p_end}
{title:Options}
{p 4 4 2}{cmdab:group} the group variable.{p_end}
{p 4 4 2}{cmdab:graph} cfactual (compares the (Depvar=0) distribution to
the (Depvar=0) distribution that would have prevailed if they had been paid like (Depvar=1)) , ufactual
(compares the (Depvar=1) distribution to the counterfactual. These will be different to the extent
that the X's of the two groups differ) and diff (difference between the how (Depvar=0) distribution
that would have prevailed if they had been paid like (Depvar=1) and (Depvar=1) distribution).{p_end}
{p 4 4 2}{cmdab:min} sets the minimum value. When omited {cmd:dfl} uses the minimum value of the specified outcome.{p_end}
{p 4 4 2}{cmdab:max} sets the maximum value. When omited {cmd:dfl} uses the maximum value of the specified outcome.{p_end}
{p 4 4 2}{cmdab:nbins} number of equaly spaced intervals.Default value 200.{p_end}
{p 4 4 2}{cmd:adaptive} produces density estimates using adaptive kernel estimation methods.
Please note that the method is slower than than the default one.{p_end}
{p 4 4 2}{cmdab:step} Sequential decomposition. This option runs N+1 models (N being the number of variables in the varlist). The first
model will consider all variables. Each subsequent model will remove one variable at the time. The final output
is a single graph with overlaying kernel density differences for the N+1 models. My impression is that if N>2 the figure
gets very clutered.{p_end}
{p 4 4 2}{cmd:gauss} specify the kernel (gaussian).{p_end}
{p 4 4 2}{cmd:epan} specify the kernel (epanechnikov).{p_end}
{p 4 4 2}{cmd:oaxaca} compute mean using the estimated density to and compare to the actual mean of log wages
or oaxaca wages (the predicted values).{p_end}
{p 4 4 2}{cmd:quietly} do not print output of propensity score estimation.{p_end}
{p 4 4 2}{cmd:probit} use probit instead of the default logit to estimate the propensity score.{p_end}
{p 4 4 2}{cmd:nxvar} assign a label to the x-axis.
{p 4 4 2}{cmd:ncfactual} assign a label to cfactual.
{p 4 4 2}{cmd:nfactual} assign a label to factual.
{p 4 4 2}{cmd:nufactual} assing a label to ufactual.
{p 4 4 2}{cmd:nogroupt} drops the group names from the figure.
{title:Examples}
{p 4 8}{stata "webuse nlsw88, clear" :. webuse nlsw88, clear}{p_end}
{p 4 8}{stata "g ttl_exp2 = ttl_exp^2" :. g ttl_exp2 = ttl_exp^2}{p_end}
{p 4 8}{stata "g lwage = log(wage)" :. g lwage = log(wage)}{p_end}
{p 4 8}{stata "dfl union ttl_exp ttl_exp2 married grade , outcome(lwage)" :. dfl union ttl_exp ttl_exp2 married grade , outcome(lwage)}{p_end}
{p 4 8}{stata "dfl union ttl_exp ttl_exp2 married grade , outcome(lwage) w(.05)" :. dfl union ttl_exp ttl_exp2 married grade , outcome(lwage) w(.05)}{p_end}
{p 4 8}{stata "dfl union ttl_exp ttl_exp2 married grade , outcome(lwage) adaptive" :. dfl union ttl_exp ttl_exp2 married grade , outcome(lwage) adaptive}{p_end}
{p 4 8}{stata "dfl union ttl_exp ttl_exp2 married grade , outcome(lwage) step(tenure collgrad)" :. dfl union ttl_exp ttl_exp2 married grade , outcome(lwage) step(tenure collgrad)}{p_end}
{title:References}
{p 4 8 2}DiNardo, J., N.M. Fortin, and T.Lemieux (1996) "Labour Market Insitutions and
the Distribution of Wages, 1973-1992: A Semiparametric Approach,"
{it:Econometrica}, 64(5): 1001-1044.{p_end}
{p 4 8 2}Van Kerm, P. (2003) "Adaptive kernel density estimation."{it:The Stata Journal}
, 3(2): 148-156.{p_end}
{title:Author}
{p 4 4 2}Joao Pedro Azevedo, World Bank, jazevedo@worldbank.org{p_end}
{title:Aknowledgements}
{p 4 4 2}This ado uses part of the code written by John DiNardo.{p_end}
{p 4 4 2}I would like to thank some Stata users who have made comments to an earlier release of this ado, in particular Jean Ries.{p_end}
{p 4 4 2}The usual disclaimer applies.{p_end}
{title:Also see}
{p 4 4 2} Manual: {hi:[R] kdensity}{p_end}
{p 4 19 2}Online: help for {help kdensity} and help for {help akdensity} and {help decompose} and {help jmp} if installed.{p_end}