svy_postestimation.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}
{* 22mar2005}{...}
{cmd:help svy postestimation}{...}
{right:also see: {helpb svy}{space 10}}
{right:{helpb svy brr}{space 6}}
{right:{helpb svy jackknife}}
{hline}

{title:Title}

{p2colset 5 33 37 2}{...}
{p2col :{hi:[SVY] svy postestimation} {hline 2}}Postestimation
tools for {cmd:svy}{p_end}
{p2colreset}{...}


{title:Description}

{pstd}
The following postestimation commands are available for {cmd:svy}:

INCLUDE help svy_postestimation


{title:Syntax for predict}

{pstd}
The syntax of {helpb predict} (and even if {cmd:predict} is allowed) following
{cmd:svy} depends upon the command used with {cmd:svy}.
Specifically, {cmd:predict} is not allowed after {cmd:svy: mean},
{cmd:svy: proportion}, {cmd:svy: ratio}, {cmd:svy: tabulate}, or
{cmd:svy: total}.


{title:Examples}

{phang}
{cmd:. webuse nhanes2}
{p_end}
{phang}
{cmd:. generate male = (sex == 1)}
{p_end}
{phang}
{cmd:. svy, subpop(male): mean zinc, over(race)}
{p_end}
{phang}
{cmd:. lincom [zinc]White - [zinc]Black}
{p_end}
{phang}
{cmd:. nlcom [zinc]White / [zinc]Black}
{p_end}

{phang}
{cmd:. webuse nhanes2e}
{p_end}
{phang}
{cmd:. generate age2 = age*age}
{p_end}
{phang}
{cmd:. svy: regress loglead age age2 female black orace region2-region4}
{p_end}
{phang}
{cmd:. nlcom peakloglead: -_b[age]/(2*_b[age2])}
{p_end}
{phang}
{cmd:. testnl -_b[age]/(2*_b[age2]) = 55}
{p_end}
{phang}
{cmd:. predictnl leadhat = exp(xb()), se(leadhat_se)}
{p_end}

{phang}
{cmd:. webuse nhanes2e}
{p_end}
{phang}
{cmd:. svy: regress loglead age female black orace region2-region4}
{p_end}
{phang}
{cmd:. test region2 region3 region4}
{p_end}
{phang}
{cmd:. test region2 region3 region4, nosvyadjust}
{p_end}
{phang}
{cmd:. test region2 region3 region4, mtest(bonferroni)}
{p_end}


{title:Also see}

{psee}
Manual:  {bf:[SVY] svy postestimation}

{psee}
Online:  {helpb svy}, {helpb svy brr}, {helpb svy jackknife};{break}
{helpb svy estat:estat},
{helpb estimates},
{helpb lincom},
{helpb mfx},
{helpb nlcom},
{helpb predict},
{helpb predictnl},
{helpb suest},
{helpb test},
{helpb testnl}
{p_end}