Ia am trying to estimate female labour supply with kernel regression
commands , after heckman twostep command applied, because I want to
compare results,
I have two equations
PARTICIPATION EQUATION
kern2 dmujoferta(Y=1 if female is working,Y=0 if not) educont(X1
years of education) expecont(X2number of years working) nhjtotal
(X3number of sons/daughters) secpublico(X4=1 if working for
Governemnt or local authorities, =0 if not),generate(parthat)
at(8.1 15 3 1) w1(.5) w2(2.6) W3(.2) W4(.0), gau
I calculated the bandwiths corresponding to the values Wi(), using
the Silverman rule of thumb with the silver command.However,i could
not estimate some kind of at() values , and I put my owns, which are
wrong .
The error I get is
invalid at
r(198)
The syntax is
kern2 varlist(min=2 max=5) [if] [in], Generate(string) AT(string)
[W1(real 0.0) W2(real 0.0) W3(real 0.0) W4(real 0.0) BIweight GAUss
RECtangle UNIform TRIangle]
AT(string) For kern2 the string contains up to 4 numbres specifying
the single covariate values of interest , and should be separated by
blanks.
Thanks in advanced,
Margarita Pastor
Univ of VAlencia
*
* 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/
