nlogit_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}
{* 20jan2005}{...}
{cmd:help nlogit postestimation}{right:dialog:  {bf:{dialog nlogit_p:predict}}}
{right:also see:  {helpb nlogit} }
{hline}

{title:Title}

{p2colset 5 34 36 2}{...}
{p2col :{hi:[R] nlogit postestimation} {hline 2}}Postestimation tools for nlogit{p_end}
{p2colreset}{...}


{title:Description}

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

{synoptset 11}{...}
{p2coldent :command}description{p_end}
{synoptline}
INCLUDE help post_estat
INCLUDE help post_estimates
INCLUDE help post_hausman
INCLUDE help post_lincom
INCLUDE help post_lrtest
INCLUDE help post_mfx
INCLUDE help post_nlcom
{synopt :{helpb nlogit postestimation##predict:predict}}predictions, residuals, influence statistics, and other diagnostic measures{p_end}
INCLUDE help post_predictnl
INCLUDE help post_test
INCLUDE help post_testnl
{synoptline}
{p2colreset}{...}


{marker predict}{...}
{title:Syntax for predict}

{p 8 16 2}
{cmd:predict} {dtype} {newvar} {ifin} [{cmd:,} {it:statistic}]

{synoptset 11 tabbed}{...}
{synopthdr :statistic}
{synoptline}
{syntab :Main}
{synopt :{opt pb}}predicted probability of choosing bottom-level, or
        choice-set, alternatives; each alternative identified by
        {it:altsetvarB}; the default{p_end}
{synopt :{opt p1}}predicted probability of choosing first-level
        alternatives; each alternative identified by {it:altsetvar1}{p_end}
{synopt :{opt p2}}predicted probability of choosing second-level
	alternatives; each choice identified by {it:altsetvar2}{p_end}
{synopt :...}{p_end}
{synopt :{opt p}{it:#}}predicted probability of choosing {it:#}-level
	alternatives; each alternative identified by {it:altsetvar#}{p_end}
{synopt :{opt xbb}}linear prediction for the bottom-level
	alternatives{p_end}
{synopt :{opt xb1}}linear prediction for the first-level
	alternatives{p_end}
{synopt :{opt xb2}}linear prediction for the second-level
	alternatives{p_end}
{synopt :...}{p_end}
{synopt :{opt xb}{it:#}}linear prediction for the {it:#}-level
	alternatives{p_end}
{synopt :{opt condpb}}Pr(each bottom alternative{c |}alternative is
	available after all earlier choices){p_end}
{synopt :{opt condp1}}Pr(each level 1 alternative) = {cmd:p1}{p_end}
{synopt :{opt condp2}}Pr(each level 2 alternative{c |}alternative is
	available after level 1 decision){p_end}
{synopt :...}{p_end}
{synopt :{opt condp}{it:#}}Pr(each level {it:#} alternative{c |}alternative is available after all earlier choices){p_end}
{synopt :{opt ivb}}inclusive value for the bottom-level
	alternatives{p_end}
{synopt :{opt iv2}}inclusive value for the second-level
	alternatives{p_end}
{synopt :...}{p_end}
{synopt :{opt iv#}}inclusive value for the {it:#}-level alternatives{p_end}
{synoptline}
{p2colreset}{...}
{p 4 6 2}
The inclusive value for the first-level alternatives is not used in
estimation; therefore, it is not calculated.{p_end}
INCLUDE help esample


{title:Options for predict}

{pstd}
Consider a nested logit model with 3 levels: Pr(ijk) = Pr(k|ij)*Pr(j|i)*Pr(i), 
then

{phang}
{opt pb}, the default, calculates the probability of choosing
bottom-level alternatives; {opt pb} = Pr(ijk).

{phang}
{opt p1} calculates the probability of choosing first-level
alternatives; {opt p1} = Pr(i).

{phang}
{opt p2} calculates the probability of choosing second-level
alternatives; {opt p2} = Pr(ij) = Pr(j|i)*Pr(i).

{phang}
{opt xbb} calculates the linear prediction for the bottom-level
alternatives.

{phang}
{opt xb1} calculates the linear prediction for the first-level
alternatives.

{phang}
{opt xb2} calculates the linear prediction for the second-level
alternatives.

{phang}
{opt condpb} = Pr(k|ij).

{phang}
{opt condp1} = Pr(i).

{phang}
{opt condp2} = Pr(j|i).

{phang}
{opt ivb} calculates the inclusive value for the bottom-level
alternatives:  {opt ivb} = ln[sum{c -(}exp(xbb){c )-]), where xbb is the
linear prediction for the bottom-level alternatives.

{phang}
{opt iv2} calculates the inclusive value for the second-level
alternatives:  {cmd:iv2} = ln[sum{c -(}exp(xb2 + tau_j*ivb){c )-}], where xb2
is the linear prediction for the second-level alternatives, ivb is the
inclusive value for the bottom-level alternatives, and tau_j are the
parameters for the inclusive value.


{title:Examples}

{phang2}{cmd:. nlogit chosen (restaurant = cost rating distance) (type = incFast incFancy kidFast kidFancy), group(family_id) nolog}{p_end}
{phang2}{cmd:. predict pb}{p_end}
{phang2}{cmd:. predict p1, p1}{p_end}
{phang2}{cmd:. predict condbp, condpb}{p_end}
{phang2}{cmd:. predict xbb, xbb}{p_end}
{phang2}{cmd:. predict xb1, xb1}{p_end}
{phang2}{cmd:. predict ivb, ivb}{p_end}


{title:Also see}

{psee}
Manual:  {bf:[R] nlogit postestimation}

{psee}
Online:  {helpb nlogit};{break}
{helpb estimates}, 
{helpb hausman}, {helpb lincom}, {helpb lrtest}, {helpb nlcom}, 
{helpb predictnl}, {helpb test}, {helpb testnl}
{p_end}