postest.hlp

是一个经济学管理应用软件 很难找的 但是经济学学生又必须用到

{smcl}
{* 07apr2005}{...}
{cmd:help postest}
{hline}

{title:Title}

{pstd}
{hi:[U] 20 Estimation and postestimation commands}


{title:Description}

{pstd}
See {help postestimation commands} for a list of Stata's postestimation
commands.   See {help estcom} for a description of estimation commands 
and their properties.

{pstd}
Estimation commands leave results in {hi:e()}.  These returned
values remain available until the next estimation command is executed.
{hi:e()} acts as a function that returns the value of the named estimation
result from the last estimation command.  You can see what is available by
typing

	{cmd:. ereturn list}

{pstd}
after an estimation command.  See {hi:[R] saved results}, 
{helpb ereturn}, and {helpb return}.

{pstd}
All estimation commands save {hi:e(sample)} indicating which observations
were used in the estimation.  This can subsequently be used with almost any
Stata command after estimation to restrict that command to the estimation
sample.


{title:Examples}

{pstd}
We use {helpb regress}; other estimation commands could also be used.

{p 8 12 2}{cmd:. regress y x1 x2 x3 if z < 5}

{pstd}
We can view the estimation matrices:

{p 8 12 2}{cmd:. matrix list e(b)}{p_end}
{p 8 12 2}{cmd:. matrix list e(V)}

{pstd}
The {helpb estat vce} command gives an alternate way of viewing the
covariance matrix:

{p 8 12 2}{cmd:. estat vce}{p_end}
{p 8 12 2}{cmd:. estat vce, corr}{p_end}
{p 8 12 2}{cmd:. estat vce, eigen}

{pstd}
We can examine the AIC and BIC information criteria.

{p 8 12 2}{cmd:. estat ic}

{pstd}
We can summarize the variables involved in the estimation command over
the observations used in the estimation.

{p 8 12 2}{cmd:. estat summarize}

{pstd}
Other postestimation commands may also be used:

{p 8 12 2}{cmd:. adjust x2 = 0 if e(sample), by(x1)}

{p 8 12 2}{cmd:. lincom 3*x1 + 500*x3}

{p 8 12 2}{cmd:. linktest}

{p 8 12 2}{cmd:. test x1=x2}

{p 8 12 2}{cmd:. test x1=x3, accum}

{p 8 12 2}{cmd:. testnl _b[x1]/_b[x2] = _b[x3]}

{p 8 12 2}{cmd:. predict r, residuals}

{pstd}
{hi:e(sample)} can be put to good use with any other Stata command after
estimation.

{p 8 12 2}{cmd:. summarize r if e(sample), detail}{p_end}
{p 8 12 2}{cmd:. summarize r if ~e(sample), detail}

{pstd}
We can even obtain predictions on different data:

{p 8 12 2}{cmd:. use newdat}{p_end}
{p 8 12 2}{cmd:. predict newy, xb}{p_end}
{p 8 12 2}{cmd:. predict newr, residuals}


{title:Also see}

{psee}
Manual:  {bf:[U] 18.9 Accessing results calculated by estimation commands},{break}
{bf:[U] 20 Estimation and postestimation commands},{break}
{bf:[U] 26 Overview of Stata estimation commands},{break}
{bf:[I] estimation commands}

{psee}
Online:  {help postestimation commands}, {help estcom};{break}
{helpb adjust},
{helpb ereturn}, {helpb estat},
{helpb estimates}, {helpb hausman}, {help level}, {helpb lincom},
{helpb linktest}, {helpb lrtest}, {helpb mfx}, {helpb nlcom}, 
{helpb predict}, {helpb predictnl},
{helpb return}, 
{helpb suest},  {helpb test}, {helpb testnl}
{p_end}