prtest.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}
{* 08dec2004}{...}
{cmd:help prtest}, {cmd:help prtestti} {right:dialogs:  {dialog prtest1:one-sample}  {dialog prtest2:two-sample}}
{right:{dialog prtestby:by() two-sample}       }
{right:immediate dialogs:  {dialog prtesti1:one-sample}  {dialog prtesti2:two-sample}}
{hline}

{title:Title}

{p2colset 5 19 21 2}{...}
{p2col :{hi:[R] prtest} {hline 2}}One- and two-sample tests of proportions{p_end}
{p2colreset}{...}


{title:Syntax}

{phang}
Test that a variable has a specified proportion

{p 8 15 2}
{cmd:prtest} {varname} {cmd:==} {it:#p} {ifin} [{cmd:,} {opt l:evel(#)}]

{phang}
Test that two variables have the same proportion

{p 8 15 2}
{cmd:prtest} {it:{help varname:varname1}} {cmd:==} {it:varname2} {ifin} [{cmd:,} {opt l:evel(#)}]

{phang}
Test that a variable has the same proportion within two groups

{p 8 15 2}
{cmd:prtest} {varname} {ifin} {cmd:,} {opth "by(varlist:groupvar)"} [{opt l:evel(#)}]

{phang}
Immediate form of one-sample test of proportion

{p 8 16 2}
{cmd:prtesti} {it:#obs1} {it:#p1} {it:#p2} 
[{cmd:,} {opt l:evel(#)} {opt c:ount}]

{phang}
Immediate form of two-sample test of proportion

{p 8 16 2}{cmd:prtesti} {it:#obs1} {it:#p1} {it:#obs2} {it:#p2} 
[{cmd:,} {opt l:evel(#)} {opt c:ount}]

{phang}
{cmd:by} may be used with {cmd:prtest} (but not with {cmd:prtesti}); see 
{helpb by}.


{title:Description}

{pstd}
{cmd:prtest} performs tests on the equality of proportions using
large-sample statistics.

{pstd}
In the first form, {cmd:prtest} tests that {varname} has a proportion of
{it:#p}.  In the second form, {cmd:prtest} tests that {it:varname1} and 
{it:varname2} have the same proportion.  In the third form, {cmd:prtest} tests
that {it:varname} has the same proportion within the two groups defined by
{it:{help varlist:groupvar}}.

{pstd}
{cmd:prtesti} is the immediate form of {cmd:prtest}; see {help immed}.

{pstd}
The {cmd:bitest} command is a better version of the first form of
{cmd:prtest} in that it gives exact p-values.  Researchers should use
{cmd:bitest} when possible, especially for small samples; see {helpb bitest}.


{title:Options}

{dlgtab:Main}

{phang}
{opth "by(varlist:groupvar)"} specifies a numeric variable that contains the
group information for a given observation.  This variable must have only two
values.  Do not confuse the {opt by()} option with the {cmd:by} prefix; both
may be specified.

{phang}
{opt level(#)} specifies the confidence level, as a percentage, for
confidence intervals.  The default is {cmd:level(95)} or as set by 
{helpb set level}.

{phang}
{opt count} specifies that integer counts instead of proportions be used in
the immediate forms of {cmd:prtest}.  In the first syntax, {cmd:prtesti}
expects that {it:#obs1} and {it:#p1} are counts{hline 2}{it:#p1} {ul:<}
{it:#obs1}{hline 2}and {it:#p2} is a proportion.  In the second syntax,
{cmd:prtesti} expects that all four numbers are integer counts, that
{it:#obs1} {ul:>} {it:#p1}, and that {it:#obs2} {ul:>} {it:#p2}.


{title:Examples}

{phang}{cmd:. prtest foreign==.4}{p_end}
{phang}{cmd:. prtest cure1==cure2}{p_end}
{phang}{cmd:. pretest cure, by(sex)}{p_end}

{phang}{cmd:. prtesti 50 .52 .70}{p_end}
{phang}{cmd:. prtesti 30 .4  45 .67}{p_end}
{phang}{cmd:. prtesti 30 4  .7, count}{p_end}
{phang}{cmd:. prtesti 30 4  45 17, count}


{title:Also see}

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

{psee}
Online:  {helpb bitest}, {helpb ci}, {helpb hotelling},
{helpb oneway}, {helpb sdtest}, {helpb signrank}, {helpb ttest}
{p_end}