help ztpflex
-------------------------------------------------------------------------------

Title

    ztpflex -- Zero-truncated Poisson mixture regression

Syntax

        ztpflex depvar [indepvars] [if] [in] [, options]

    where depvar has to be a strictly postive outcome.

    options               Description
    -------------------------------------------------------------------------
    irr                   report incidence-rate ratios
    nonadaptive           use standard Gauss-Hermite quadrature; default is
                            adaptive quadrature
    intpoints(#)          choose the number of quadrature points used for the
                            approximation; default is intpoints(30)
    noconstant            suppress constant term
    vce(vcetype)          vcetype may be oim, robust, cluster clustvar, or
                            opg
    vuong                 perform Vuong test of ztpflex versus ztnbp
    maximize_options      control the maximization process; see [R] maximize
    -------------------------------------------------------------------------
    bootstrap and jackknife are allowed; see prefix.


Description

    ztpflex fits a zero-truncated Poisson model with a more flexible mixing
    distribution than ztpnm. The integral is approximated using adaptive
    Gauss-Hermite quadrature.  Generally, a higher number of quadrature
    points leads to a more accurate approximation, but it takes longer to
    converge. It is highly recommended to check the sensitivity of the
    results; see [XT] quadchk.

    ztpflex, nonadaptive uses standard Gauss-Hermite quadrature.  Generally,
    this method is less accurate even if the number of quadrature points is
    high.

    This program uses ml lf method.


Options

    irr reports incidence-rate ratios.

    nonadaptive uses standard Gauss-Hermite quadrature; the default is
        adaptive quadrature.

    intpoints(#) chooses the number of points used for the approximation.
        The default is intpoints(30). The maximum is 195. Generally, a higher
        number of points leads to a more accurate approximation, but it takes
        longer to converge.  It is highly recommended to check the
        sensitivity of the results.

    noconstant suppresses the constant term (intercept) in the model.

    vce(vcetype) specifies the type of standard error reported, which
        includes types that are derived from asymptotic theory, that are
        robust to some kinds of misspecification, and that allow for
        intragroup correlation; see [R] vce_option.

    vuong performs a Vuong test of ztpflex versus ztnbp.

    maximize_options: difficult, technique(algorithm_spec), iterate(#),
        [no]log, trace, gradient, showstep, hessian, showtolerance,
        tolerance(#), ltolerance(#), nrtolerance(#), nonrtolerance, and
        from(init_specs); see [R] maximize.  These options are seldom used.

        difficult is the default.

Remarks for quadcheck

    See Remarks in [XT] quadchk.

Author

    Helmut Farbmacher
    Munich Center for the Economics of Aging (MEA)
    Max Planck Society, Germany
    farbmacher@mea.mpisoc.mpg.de

Reference

    Farbmacher, H. 2012: Extensions of hurdle models for overdispersed count
             data, Health Economics, forthcoming.

       Help:  [R] ztp, [R] ztpnm, [R] ztnbp