[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: Bootstrapping standard errors from two different regression
Hi all, I have a basic question about using the
bootstrap command for a case that cannot be addressed
by the manual's coverage of bootstrap:
Suppose I have two separate regressions from two
different datasets, yielding the vectors b_1 and b_2,
respectively. The parameter of interest is b_star =
b_1k * b_2j, where k and j respectively refer to one
of the coefficients in each of the vectors.
I want to get the standard error of the estimator
b_star, which presumably can only be done by
bootstrapping. Expressing my question more generally:
How do I get inference statistics for parameters which
are functions of other parameters that were estimated
in different regressions?
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
* For searches and help try: