Huber correction for two-stage least squares
--------------------------------------------

^hreg2sls^  depvar [varlist1 [(varlist2)]] [weight] [^if^ exp] [^in^ range] ]
                [^,^ ^g^roup^(^varname^)^ ^l^evel^(^#^)^ regress_options ]


^pweight^s, ^aweight^s, ^fweight^s, and ^iweight^s are allowed; see help @weights@.
All commands share the features of estimation commands; see [4] @est@.
To reset problem-size limits, see help @matsize@.

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

^hreg2sls^ estimates a two-stage least squares regression with robust standard
errors based on Huber's formula for individual-level data.  Consistent standard
errors are estimated even if there is heteroscedasticity, clustered sampling,
or the data are weighted.  ^hlogit^, ^hprobit^, and ^hereg^, do the same for
logit, probit, and exponential regression models.

This command is identical to @hreg@ except that ^hreg2sls^ permits
instrumental variable estimation.

Options
-------

^group(^varname^)^ identifies the cluster to which an observation belongs.
        Cases with the same value of varname are assumed to be drawn from
        the same cluster.  The data need not be sorted by varname.

regress_options, logit_options, probit_options, and ereg_options are any of
        the options allowed by ^regress^, ^logit^, ^probit^, or ^ereg^; see
        help @regress@, help @logit@, help @probit@, or help @ereg@.

Example
-------

 . ^hreg2sls y1 y2 x1 x2 x3 (x1 x2 x3 z1 z2), group(cluster)^

Also see
--------

 manual:  [5s] huber, [5s] regress
On-line:  help for @hreg@, @correlate@, @deff@