help for glgamma2                                                (Joseph Hilbe)

Generalized 2-parameter log-gamma regression

glgamma2 depvar [varlist] [if exp] [in range] [, offset(varname) exposure(varname) cluster(varname) level(#) from(asis) eform robust nolog maximize_options survey_options]

aweights, fweights, iweights, and pweights are allowed; see help weights.

glgamma2 provides access to all maximize options; see help maximize.

glgamma2 provides access to all survey options; see help svy.


glgamma2 fits a maximum-likelihood generalized 2-parameter log-gamma regression model of depvar on indepvars, where depvar is a non-negative count variable. The program may be used to model under-dispersed Poisson count data. Under-dispersion is indicated if the scale parameter, phi, is greater than 1; values under 1 indicate over-dispersion. Other methods are available for modeling overdispersed Poisson data, including the negative binomial, but few methods are commonly available to deal with under-dispersion. glgamma2 can also be used with models having a positive continuous response.

glgamma2 allows the parameterization of the natural log of the scale parameter, lnphi. Without parameterization of lnphi, the model reduces to the non-generalized form, but without the additional transformed value of phi. Run lgamma2 if phi is not to be parameterized.

glgamma2 acccepts all of the help maximize options, the constraint() option, and all survey options and capabilities documented in [SVY]; including multi-level surveys; poststratification; and BRR, jackknife, and linearization VCE estimators.

This program uses ml lf method.


+-------+ ----+ Model +------------------------------------------------------------

offset(varname) specifies a varname in model with coefficient constrained to 1.

exposure(varname) specifies a ln(varname) in model with coefficient constrained to 1.

constraints(constraints) apply specified linear constraints.

+-----------+ ----+ SE/Robust +--------------------------------------------------------


robust specifies that the Huber/White/sandwich estimator of variance is to be used in place of the traditional calculation. robust combined with cluster() allows observations which are not independent within cluster (although they must be independent between clusters). If you specify pweights, robust is implied.

vce(options) allowed. vce() supports robust, opg, and native. vce does not support options bootstrap or jacknife, However, glgamma2 does support the bootstrap and jacknife commands, so these modeling capabilities are allowed.

+-----------+ ----+ Reporting +-------------------------------------------------------- level(#) specifies the confidence level, in percent, for confidence intervals of the coefficients; see help level.

nolog suppresses the iteration log.

+-------------+ ----+ max options +------------------------------------------------------

maximize_options: technique(algorithm_spec), [no]log, trace, hessian, gradient, showstep, shownrtolerance, difficult, iterate(#), tolerance(#), ltolerance(#), gtolerance(#), nrtolerance(#), nonrtolerance, from(init_specs); see maximize.

+-------------+ ----+ svy options +------------------------------------------------------

survey_options are all available. See help svy

Author and support

Joseph Hilbe, Arizona State University: jhilbe@aol.com


glgamma2 is a user authored program. Support is by author.

glgamma2 requires a response with any positive real number. A response of 0 will result in an error.


. glgamma2 los hmo white type2 type3, nolog lnphi(hmo white type2 type3)

. glgamma2 los hmo white type2 type3, nolog exposure(pop) cluster(state) lnphi(hmo white)

. bootstrap: glgamma2 los hmo white type2 type3, nolog lnphi(hmo white type2 type3)

Also see

Reference: Hardin, J. and Hilbe, J., (2001), Generalized Linear Models and Extensions, Stata Press.

Online: help streg