You should be using
. reg lnwage sex age agesq [pweight=calwght]
if you have no other survey design info (otherwise see -help svy-).
See -help weights- for the following: "pweights, or sampling weights,
are weights that denote the inverse of the probability that the
observation is included due to the sampling design," which implies
that in the absence of any complex sample design, they represent the
number of people in the group that a given obs was drawn from (i.e. if
the probability is 1/N then the inverse of the probability is N). You
should certainly never use fweights or aweights with survey data
unless you care not at all about getting correct standard errors, test
statistics, etc.
RTSM:
You might also read the [SVY] manual before you spend too much time estimating.
On 2/6/06, Ada Ma <[email protected]> wrote:
> I'm estimating a mincer wage equation:
> The data set I'm using comes with a population weight variable, so I estimate:
> reg lnwage sex age agesq [fweight=calwght]
*
* 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/