hadimvo.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}
{* 29mar2005}{...}
{cmd:help hadimvo}{right:also see:  {help undocumented}}
{hline}

{title:Title}

{p 4 17 2}
{hi:hadimvo} {hline 2} Identify multivariate outliers


{title:Syntax}

{p 4 12 2}{cmd:hadimvo} {varlist} {ifin}{cmd:,} {cmdab:g:enerate:(}{it:newvar1} [{it:newvar2}]{cmd:)}
[{opt p(#)}]


{title:Description}

{pstd}
{cmd:hadimvo} identifies multiple outliers in multivariate data using the
method of Hadi (1992, 1994), creating {it:newvar1} equal to 1 if an
observation is an "outlier" and 0 otherwise.  Optionally, {it:newvar2} can
also be created containing the distances from the basic subset.


{title:Options}

{phang}{cmd:generate(}{it:newvar1} [{it:newvar2}]{cmd:)} is not optional; it
identifies the new variable(s) to be created.  Whether you specify two
variables or one, however, is optional.  {it:newvar2}, if specified, will
contain the distances (not the distances squared) from the basic subset.  That
is, specifying {cmd:gen(odd)} creates the variable {hi:odd} containing 1 if the
observation is an outlier in the Hadi sense and 0 otherwise.  Specifying
{cmd:gen(odd dist)} also creates variable {hi:dist} containing the Hadi
distances.

{phang}{cmd:p(}{it:#}{cmd:)} specifies the significance level for outlier
cutoff; 0 < {it:#} < 1.  The default is {cmd:p(.05)}.  Larger numbers identify
a larger proportion of the sample as outliers.  If {it:#} is specified greater
than 1, it is interpreted as a percent.  Thus, {cmd:p(5)} is the same as
{cmd:p(.05)}.


{title:Examples}

    {cmd:. hadimvo price weight, gen(odd)}
    {cmd:. list if odd}{right:/* list the outliers            */  }
    {cmd:. summ price weight if ~odd}{right:/* summary stats for clean data */  }

    {cmd:. drop odd}
    {cmd:. hadimvo price weight, gen(odd D)}
    {cmd:. summarize D, detail}
    {cmd:. sort D}
    {cmd:. list make price weight D odd}

    {cmd:. hadimvo price weight mpg, gen(odd2 D2) p(.01)}
    {cmd:. regress} {it:...} {cmd:if ~odd2}


{title:Also see}

{psee}
Online:  {helpb mvreg}, {helpb regress},
{help regress postestimation}, {helpb sureg}
{p_end}