用stata建立command去比较pooled ols, FE和 RE estimator

stimator comparison

在stata中，我们会用以下command 去做estimator的对比

Estimator comparison

* Compare various estimators (with cluster-robust se.s)

. global xlist exp exp2 wks ed

. quietly regress lwage $xlist, vce(cluster id)

. estimates store OLS

. quietly xtgee lwage exp exp2 wks ed, corr(ar 2)

vce(robust)

. estimates store PFGLS

. quietly xtreg lwage $xlist, be

. estimates store BE

. quietly xtreg lwage $xlist, re vce(robust)

. estimates store RE

. quietly xtreg lwage $xlist, fe vce(robust)

. estimates store FE

. estimates table OLS PFGLS BE RE FE, b(%9.4f) se stats(N)

但是在这段command中的最后一行，b(%9.4f) se stats(N) 是什么意思？？？求指点和帮助！谢谢！o

A. Colin Cameron Univ. of California - Davis (Based on A. Colin Cameron and Praveconometrics using Stat

help estimates table里有。就是几个控制表格显示格式的选项。

夏目贵志 发表于 2017-5-1 05:41
help estimates table里有。就是几个控制表格显示格式的选项。

可不可以具体说一下：b(%9.4f) se stats(N) 里面的%9.4f是什么意思？？？

猫咪祧 发表于 2017-5-2 03:30
可不可以具体说一下：b(%9.4f) se stats(N) 里面的%9.4f是什么意思？？？

看帮助或manual里面的解释

Title

    [R] estimates table -- Compare estimation results


Syntax

        estimates table [namelist] [, options]


    where namelist is a name, a list of names, _all, or *.
        A name may be ., meaning the current (active) estimates.
        _all and * mean the same thing.


    options                  Description
    ----------------------------------------------------------------------------------------------------------
    Main
      stats(scalarlist)      report scalarlist in table
      star[(#1 #2 #3)]       use stars to denote significance levels

    Options
      keep(coeflist)         report coefficients in order specified
      drop(coeflist)         omit specified coefficients from table
      equations(matchlist)   match equations of models as specified

    Numerical formats
      b[(%fmt)]              how to format coefficients, which are always reported
      se[(%fmt)]             report standard errors and use optional format
      t[(%fmt)]              report t or z and use optional format
      p[(%fmt)]              report p-values and use optional format
      stfmt(%fmt)            how to format scalar statistics

    General format
      varwidth(#)            use # characters to display variable names and statistics
      modelwidth(#)          use # characters to display model names

      eform                  display coefficients in exponentiated form
      varlabel               display variable labels rather than variable names
      newpanel               display statistics in separate table from coefficients

      style(oneline)         put vertical line after variable names; the default
      style(columns)         put vertical line separating every column
      style(noline)          suppress all vertical lines

      coded                  display compact table

    Reporting
      display_options        control row spacing, line width, and display of omitted variables and base and
                               empty cells

      title(string)          title for table
    ----------------------------------------------------------------------------------------------------------