tetrachoric.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}
{* 19mar2005}{...}
{cmd:help tetrachoric} {right:dialog:  {bf:{dialog tetrachoric}}}
{hline}

{title:Title}

{p2colset 5 24 26 2}{...}
{p2col :{hi:[R] tetrachoric} {hline 2}}Tetrachoric correlations for binary
	variables{p_end}
{p2colreset}{...}


{title:Syntax}

{p 8 20 2}
{cmd:tetrachoric} {varlist} {ifin} {weight}
[{cmd:,} {it:options}]

{synoptset 16 tabbed}{...}
{synopthdr}
{synoptline}
{syntab:Main}
{synopt:{opt avail:able}}compute pairwise correlations 
	using all available data{p_end}
{synopt:{opth for:mat(%fmt)}}display format for correlations; 
	default {cmd:%8.4g}{p_end}
{synopt:{opt notab:le}}suppress display of correlations{p_end}
{synopt:{opt pos:def}}modify correlation matrix to be positive
	(semi)definite{p_end}
{synoptline}
{p2colreset}{...}
{p 4 6 2}
{cmd:by} may be used with {cmd:tetrachoric}; see {helpb by}.{p_end}
{p 4 6 2}
{cmd:fweight}s are allowed; see {help weight}.


{title:Description}

{pstd}
{cmd:tetrachoric} computes estimates of the tetrachoric correlation coefficients
of the variables in {varlist}.  All these variables should be 0, 1, or
{help missing} valued.

{pstd}
Tetrachoric correlations assume a latent bivariate normal distribution 
({it:X1},{it:X2}) for each pair of variables ({it:v1},{it:v2}), with a
threshold model for the manifest variables, {bind:{it:vi} = 1} if and only 
if {bind:{it:Xi} > 0}.  The means and variances of the latent variables are 
not identified, but the correlation {it:r} of {it:X1} and {it:X2} can be 
estimated from the joint distribution of {it:v1} and {it:v2}.  These are 
called the tetrachoric correlation coefficients.

{pstd}
{cmd:tetrachoric} computes the pairwise estimates of the tetrachoric
correlations by Edwards & Edwards.

{pstd}
The pairwise correlation matrix is returned as {cmd:r(corr)} and can be
used to perform a factor or a principal component analysis of binary
variables using {helpb factormat} or {helpb pcamat}.


{title:Options}

{dlgtab:Main}

{phang}{opt available}
computes the tetrachoric correlation of two binary variables {it:var1}
and {it:var2} using all observations that are not missing on {it:var1}
and {it:var2}.  The default is to use the observations that are not
missing for all variables in {it:varlist}.

{phang}{opth format(%fmt)}
specifies the display format for correlations.
Default: {cmd:format(%8.4g)}.

{phang}{opt notable}
suppresses the display of the correlation matrix. 

{phang}{opt posdef}
modifies the correlation matrix so that it is positive semidefinite, i.e.,
a proper correlation matrix.  The modified result is the correlation matrix
associated with the least squares approximation of the tetrachoric
correlation matrix by a positive-semidefinite matrix.


{title:Examples}

    {cmd:. tetrachoric item*}
    {cmd:. matrix C = r(corr)}
    {cmd:. pcamat C, fac(2)}
    
    
{title:Also see}

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

{psee}
Online:  {helpb correlate},
{helpb factor},
{helpb ktau},
{helpb pca},
{help tabulate twoway}
{p_end}