Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: Mkspline, age classes and probit


From   "Elena Giarda" <elena.giarda@prometeia.it>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: Mkspline, age classes and probit
Date   Tue, 16 Sep 2008 10:43:22 +0200

Dear all,

I'm fitting a probit/logit model on Italian labour force micro data where the dependent variable is "female labour market participation" and among the explicative variables I have age, education level, dummy variables for the presence of children of various ages, partner's education level and a dummy for an unemployed partner.

My original idea was to estimate the relevance of cohort effects, but after 2004 (my data date back to 1993) the variable age has not been made available anymore, if not grouped in 5-year age classes, making it impossible to create birth cohorts. In order to study the effects of age on female participation I thought of using age splines, but I'm not sure it makes sense to build them with age classes. If this is reasonable, would the interpretation of the estimated coefficients be the standard one of the probit/logit specification?

Many thanks!
Elena


*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/




© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index