rreg.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}
{* 21mar2005}{...}
{cmd: help rreg}{right:dialog:  {bf:{dialog rreg}} {space 14}}
{right: also see:  {help rreg postestimation}}
{hline}

{title:Title}

{p2colset 5 17 19 2}{...}
{p2col :{hi:[R] rreg} {hline 2}}Robust regression{p_end}
{p2colreset}{...}


{title:Syntax}

{p 8 14 2}
{cmd:rreg} {depvar} [{indepvars}] {ifin} 
[{cmd:,} {it:options}] 

{synoptset 22 tabbed}{...}
{synopthdr:options}
{synoptline}
{syntab:Model}
{synopt :{opt tu:ne(#)}}use {it:#} as the biweight tuning constant; default is
{cmd:tune(7)}{p_end}

{syntab:Reporting}
{synopt :{opt l:evel(#)}}set confidence level; default is {cmd:level(95)}
{p_end}
{synopt :{opth g:enwt(newvar)}}create {newvar} containing the weights assigned
to each observation{p_end}

{syntab:Opt options}
{synopt :{it:{help rreg##optimize_options:optimization_options}}}control the optimization process; seldom used{p_end}
{synopt :{opt g:raph}}graph weights during convergence{p_end}
{synoptline}
{p2colreset}{...}
{p 4 6 2}{it:depvar} and {it:indepvars} may contain time-series operators; see 
{help tsvarlist}.{p_end}
{p 4 6 2}{cmd:by}, {cmd:rolling}, {cmd:statsby}, and {cmd:xi} are allowed; see
{help prefix}.{p_end}
{p 4 6 2}See {help rreg postestimation} for additional capabilities of estimation commands.


{title:Description}

{pstd}
{opt rreg} performs one version of robust regression of {depvar} on
{indepvars}.

{pstd}
Also see {bf:[R] regress} for standard regression with robust variance
estimates and {bf:[R] qreg} for quantile (including median or
least-absolute-residual) regression.


{title:Options}

{dlgtab:Model}

{phang}
{opt tune(#)} is the biweight tuning constant.  The default is 7, meaning 
7 times the median absolute deviation from the median residual (MAD).  Lower
tuning constants downweight outliers rapidly but may lead to unstable
estimates (below 6 is not recommended).  Higher tuning constants produce
milder downweighting.

{dlgtab:Reporting}

{phang}
{opt level(#)}; see {help estimation options##level():estimation options}. 

{phang}
{opth genwt(newvar)} creates the new variable {it:newvar} containing the
weights assigned to each observation.

{marker optimize_options}{...}
{dlgtab:Opt options}

{phang}
{it:optimization_options}: 
{opt iter:ate(#)}, {opt tol:erance(#)}, [{cmd:{ul:no}}]{cmd:{ul:lo}}{cmd:g}.
{opt iterate()} specifies the maximum number of iterations; iterations stop
when the maximum change in weights drops below {opt tolerance()}; and
{opt log}/{opt nolog} specifies whether or not to show the iteration log.
These options are seldom used.

{phang}
{opt graph} allows you to graphically watch the convergence of the
iterative technique.  The weights obtained from the most recent round of
estimation are graphed against weights obtained from the previous round.


{title:Examples}

    {cmd:. rreg mpg weightd weightf foreign}{right:(fit model)                }
    {cmd:. test weightd==weightf}{right:(test hypothesis)          }
    {cmd:. predict predmpg}{right:(obtain predicted values)  }
    {cmd:. predict r, resid}{right:(obtain residuals)         }
    {cmd:. rreg}{right:(redisplay results)        }

    {cmd:. rreg mpg weightd weightf foreign, gen(wt)}{right:(fit model)                }
    {cmd:. predict hat, hat}{right:(obtain corrected hat)     }


{title:Also see}

{psee}
Manual:  {bf:[R] rreg}

{psee}
Online:  {help rreg postestimation};{break} 
{helpb qreg}, {helpb regress}, {help regress postestimation}{p_end}