请问predict(pu0)的含义是什么呢？谢谢！

Its similar to atmeans under logistic.  For random effect, we need to fix the random variable, 0 is as mean of the random variable, I think.

Title

    [XT] xtlogit postestimation -- Postestimation tools for xtlogit


Description

    The following postestimation commands are available after xtlogit:

    Command          Description
    ---------------------------------------------------------------------------------------------------------------------
        contrast     contrasts and ANOVA-style joint tests of estimates
    (1) estat        AIC, BIC, VCE, and estimation sample summary
        estimates    cataloging estimation results
        hausman      Hausman's specification test
        lincom       point estimates, standard errors, testing, and inference for linear combinations of coefficients
        lrtest       likelihood-ratio test
    (2) margins      marginal means, predictive margins, marginal effects, and average marginal effects
        marginsplot  graph the results from margins (profile plots, interaction plots, etc.)
        nlcom        point estimates, standard errors, testing, and inference for nonlinear combinations of coefficients
        predict      predictions, residuals, influence statistics, and other diagnostic measures
        predictnl    point estimates, standard errors, testing, and inference for generalized predictions
        pwcompare    pairwise comparisons of estimates
        test         Wald tests of simple and composite linear hypotheses
        testnl       Wald tests of nonlinear hypotheses
    ---------------------------------------------------------------------------------------------------------------------
    (1) estat ic is not appropriate after xtlogit, pa.
    (2) The default prediction statistic for xtlogit, fe, pu1, cannot be correctly handled by margins; however, margins
        can be used after xtlogit, fe with the predict(pu0) option or the predict(xb) option.


Syntax for predict

    Random-effects model

        predict [type] newvar [if] [in] [, RE_statistic nooffset]

    Fixed-effects model

        predict [type] newvar [if] [in] [, FE_statistic nooffset]

    Population-averaged model

        predict [type] newvar [if] [in] [, PA_statistic nooffset]

    RE_statistic   Description
    ---------------------------------------------------------------------------------------------------------------------
    Main
      xb           linear prediction; the default
      pu0          probability of a positive outcome assuming the random effect is zero
      stdp         standard error of the linear prediction
    ---------------------------------------------------------------------------------------------------------------------

    FE_statistic   Description
    ---------------------------------------------------------------------------------------------------------------------
    Main
      pc1          probability of a positive outcome conditional on one positive outcome within group; the default
      pu0          probability of a positive outcome assuming the fixed effect is zero
      xb           linear prediction
      stdp         standard error of the linear prediction
    ---------------------------------------------------------------------------------------------------------------------

    PA_statistic   Description
    ---------------------------------------------------------------------------------------------------------------------
    Main
      mu           probability of depvar; considers the offset()
      rate         probability of depvar
      xb           linear prediction
      stdp         standard error of the linear prediction
      score        first derivative of the log likelihood with respect to xb
    ---------------------------------------------------------------------------------------------------------------------

    These statistics are available both in and out of sample; type predict ... if e(sample) ... if wanted only for the
      estimation sample.
    The predicted probability for the fixed-effects model is conditional on there being only one outcome per group.  See
      [R] clogit for details.

蓝色 你好， 我想请教您一个关于在"Panel logit model with random effects" 模型中计算边际效应的问题。 请问在STATA 中计算这个模型 (xtlogit..., re)的边际效应的程序是: margins, dydx(*) predict(pu0) 吗? 这个在STATA手册的什么地方可以找到? 期待您的答复 很急

手册上没有这么具体到每个命令的
i先看margins的部分，看里面的例子，许多都是相似的，然后在看xtlogit命令相关的内容

遇到了同样的问题，模型回归系数和边际效应结果一样，用了margins,dydx(*)  predict(pu0)命令后果然不一样了，但是不知道predict（pu0）的含义？

麻烦问题以下如何输出logit回归结果的几率比和多个统计量呢？谢谢~

Good answer！

请问楼主解决这个问题了吗？我看到陈强高级那本书里说logit随机效应不能求边际效应，不知道怎么搞的

你好，我想请问一下，我用xtlogit回归得到的系数是显著的，用这个margins,dydx(*)  predict(pu0)命令得到的边际效应，后边的p值不显著可以吗

你好，我想请问一下，我用xtlogit回归得到的系数是显著的，用这个margins,dydx(*)  predict(pu0)命令得到的边际效应，后边的p值不显著可以吗

你好，我想请问一下，我用xtlogit回归得到的系数是显著的，用这个margins,dydx(*)  predict(pu0)命令得到的边际效应，后边的p值不显著可以吗

您好，我也遇到了这个问题，请问边际效应不显著，是怎么处理的呢