.-
help for ^areg2^                                               
.-

2SLS with a large dummy-variable set
------------------------------------

    ^areg2^ depvar [^if^ exp] [^in^ range]^, ^exog^([exogvars]) ^endog^([endogvars]) ^iv^([insvars]) ^fe^([fe variable]) [^choice^([string of options normally admissible for areg]) ^nofsfe^ ]^

no weights are currently allowed

^areg2^ shares the features of all estimation commands; see help @est@.

^areg2^, typed without arguments, does not redisplay previous estimation results.

To reset problem-size limits, see help @matsize@.


Description
-----------

^areg2^ estimates 2SLS linear regression absorbing one categorical factor.



Options
-------

^fe(^varname^)^ is not optional; it specifies the categorical variable which
    is to be included in the regression as if it were specified by dummy
    variables.

^choice^ allows you to insert a string of options that would normally
    come after the comma when you call @areg@ in stata.  An example
    includes:

	^choice^(robust)

which would specify that the Huber/White/sandwich estimator of variance is to 
    be used in place of the traditional calculation.

See the manual on @areg@ to learn more about these options.

^nofsfe^ allows you to run the first-stage regressions without fixed
effects.  My suspicion is that in cases of perfect multicolinearity
between the instruments and the fixed effects, the fixed effects are
effectively dropped anyway, but this option lets you make the decision
explicitly.


^Warning:^ with many endogenous variables, ^areg2^ can be very slow --
it has to run N+1 separate fixed effects regressions.

Examples
--------

	. ^areg2 depvar, exog( price) endog( weight) iv( length) fe(rep78)^

	. ^areg2 depvar, exog(lofstay) endog(occrate) iv(region2-region4) fe(dcode) choice(robust)^

Also see
--------

 Manual:  ^[U] 26 Estimation and post-estimation commands^
          ^[U] 35 Overview of model estimation^
	  ^[R] areg^
On-line:  help for @est@; @lincom@, @predict@, @regress@, @test@, @testnl@, @vce@, @xtreg@