pkequiv.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 pkequiv}{right:dialog:  {bf:{dialog pkequiv}}}
{hline}

{title:Title}

{p2colset 5 20 22 2}{...}
{p2col :{hi:[R] pkequiv} {hline 2}}Perform bioequivalence tests{p_end}
{p2colreset}{...}


{title:Syntax}

{p 8 16 2}
{cmd:pkequiv} {it:outcome treatment period sequence id} {ifin} [{cmd:,}
{it:options}]

{synoptset 19 tabbed}{...}
{synopthdr}
{synoptline}
{syntab :Options}
{synopt :{opt com:pare(string)}}compare the two specified values of the
treatment variable{p_end}
{synopt :{opt li:mit(#)}}equivalence limit (between .10 and .99); default is
.2{p_end}
{synopt :{opt le:vel(#)}}set confidence level; default is
{cmd:level(90)}{p_end}
{synopt :{opt fie:ller}}calculate confidence interval by Fieller's theorem{p_end}
{synopt :{opt sym:metric}}calculate symmetric equivalence interval{p_end}
{synopt :{opt ander:son}}Anderson and Hauck hypothesis test for
bioequivalence{p_end}
{synopt :{opt tost}}two one-sided hypothesis tests for bioequivalence{p_end}
{synopt :{opt nob:oot}}do not estimate probability that CI lies within
confidence limits{p_end}
{synoptline}
{p2colreset}{...}


{title:Description}

{pstd}
{cmd:pkequiv} performs bioequivalence testing for two treatments.  By
default, {cmd:pkequiv} calculates a standard confidence interval symmetric
about the difference between the two treatment means.  {cmd:pkequiv} also
calculates confidence intervals symmetric about zero and intervals based on
Fieller's theorem.  Additionally, {cmd:pkequiv} can perform interval
hypothesis tests for bioequivalence.

{pstd}
{cmd:pkequiv} is one of the pk commands.  Please read {helpb pk} before reading
this entry.


{title:Options}

{dlgtab:Options}

{phang}
{opt compare(string)} specifies the two treatments to be tested for
equivalence.  In some cases, there may be more than two treatments, but the
equivalence can only be determined between any two treatments.

{phang}
{opt limit(#)} specifies the equivalence limit.  The default is .2.  The
equivalence limit can only be changed symmetrically; that is, it is not
possible to have a .15 lower limit and a .2 upper limit in the same test.

{phang}
{opt level(#)} specifies the confidence level, as a percentage, for confidence
intervals.  The default is {cmd:level(90)}.  Note that this is not controlled
by the {cmd:set level} command.

{phang}
{opt fieller} specifies that an equivalence interval based on Fieller's
theorem be calculated.

{phang}
{opt symmetric} specifies that a symmetric equivalence interval be calculated.

{phang}
{opt anderson} specifies that the Anderson and Hauck hypothesis test for
bioequivalence be computed.  This option is ignored when calculating
equivalence intervals based on Fieller's theorem or when calculating a
confidence interval that is symmetric about zero.

{phang}
{opt tost} specifies that the two one-sided hypothesis tests for
bioequivalence be computed.  This option is ignored when calculating
equivalence intervals based on Fieller's theorem or when calculating a
confidence interval that is symmetric about zero.

{phang}
{opt noboot} prevents the estimation of the probability that the confidence
interval lies within the confidence limits.  If this option is not specified,
this probability is estimated by resampling the data.


{title:Remarks}

{pstd}
{cmd:pkequiv} is designed to conduct tests for bioequivalence based on data
from a crossover experiment.  {cmd:pkequiv} requires that the user
specify the {it:outcome}, {it:treatment}, {it:period}, {it:sequence}, and
{it:id} variables.  The data must be in the same format as produced by
{helpb pkshape}.


{title:Examples}

{phang}{cmd:. pkequiv outcome treat period seq id}{p_end}

{phang}{cmd:. pkequiv outcome treat period seq id, symmetric}{p_end}

{phang}{cmd:. pkequiv outcome treat period seq id, tost anderson}


{title:Also see}

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

{psee}
Online:  {helpb pk}, {helpb pkcollapse}, {helpb pkcross},
{helpb pkexamine}, {helpb pkshape}, {helpb pksumm}, {helpb statsby}
{p_end}