stata10里面有  xtmelogit

Title

    [XT] xtmelogit -- Multilevel mixed-effects logistic regression

Syntax

        xtmelogit depvar [fe_equation] || re_equation [|| re_equation ...] [, options]

    and where the syntax of fe_equation is

            indepvars [if] [in] [, fe_options]

    and the syntax of re_equation is one of:

        for random coefficients

            levelvar: [varlist] [, re_options]

        for a random effect among the levels of a factor variable

            levelvar: R.varname [, re_options]

    where levelvar is the grouping variable for the random effects at that level, or _all for the inclusive group
    comprising all observations.

    fe_options                      description
    --------------------------------------------------------------------------------------------------------------
    Model
      noconstant                    suppress the constant from the fixed-effects equation
      offset(varname)               include varname in model with coefficient constrained to 1
    --------------------------------------------------------------------------------------------------------------

    re_options                      description
    --------------------------------------------------------------------------------------------------------------
    Model
      covariance(vartype)           variance-covariance structure of the random effects
      noconstant                    suppress the constant from the random-effects equation
      collinear                     keep collinear variables
    --------------------------------------------------------------------------------------------------------------

    options                         description
    --------------------------------------------------------------------------------------------------------------
    Model
      binomial(varname|#)           set binomial trials if data are in binomial form

    Integration
      laplace                       use Laplacian approximation; equivalent to intpoints(1)
      intpoints(# [# ...])          set the number of integration (quadrature) points; default is 7

    Reporting
      level(#)                      set confidence level; default is level(95)
      or                            report fixed-effects coefficients as odds ratios
      variance                      show random-effects parameter estimates as variances and covariances
      noretable                     suppress random-effects table
      nofetable                     suppress fixed-effects table
      estmetric                     show parameter estimates in the estimation metric
      noheader                      suppress output header
      nogroup                       suppress table summarizing groups
      nolrtest                      do not perform LR test comparing to logistic regression

    Max options
      maximize_options              control the maximization process during gradient-based optimization; seldom
                                      used
      retolerance(#)                tolerance for random-effects estimates; default is retolerance(1e-8); seldom
                                      used
      reiterate(#)                  maximum number of iterations for random-effects estimation; default is
                                      reiterate(50); seldom used
      matlog                        parameterize variance components using matrix logarithms
      refineopts(maximize_options)  control the maximization process during refinement of starting values
    --------------------------------------------------------------------------------------------------------------

    vartype                  description
    --------------------------------------------------------------------------------------------------------------
    independent              one variance parameter per random effect, all covariances zero; the default unless a
                               factor variable is specified
    exchangeable             equal variances for random effects, and one common pairwise covariance
    identity                 equal variances for random effects, all covariances zero; the default for factor
                               variables
    unstructured             all variances-covariances distinctly estimated
    --------------------------------------------------------------------------------------------------------------
    indepvars and varlist may contain time-series operators; see tsvarlist.
    bootstrap, by, jackknife, rolling, statsby, and xi are allowed; see prefix.
    See [XT] xtmelogit postestimation for features available after estimation.

Description

    xtmelogit fits mixed-effects models for binary/binomial responses.  Mixed models contain both fixed effects
    and random effects.  The fixed effects are analogous to standard regression coefficients and are estimated
    directly.  The random effects are not directly estimated (although they may be obtained postestimation) but
    are summarized according to their estimated variances and covariances.  Random effects may take the form of
    either random intercepts or random coefficients, and the grouping structure of the data may consist of
    multiple levels of nested groups.  The distribution of the random effects is assumed to be Gaussian.  The
    conditional distribution of the response given the random effects is assumed to be Bernoulli, with success
    probability determined by the logistic cumulative distribution function (c.d.f.).  Since the log likelihood
    for this model has no closed form, it is approximated by adaptive Gaussian quadrature.

感谢你的帮助！不过好像不是这个模型，有人建议在stata10里面用

net search mixlogit

net install mixlogit

但之后我就不知道怎么操作了

安装完看帮助

里面一般是有例子的

十分感谢！

在你的帮助之下我已经把模型做出来了，不过不懂看结果，不知道怎么样的标准表示这个模型好，望指点！

不懂看结果，可能是你对理论不了解

软件只是实现理论的一个工具。

你去看看相关的理论的书。