centile.hlp

来自「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」· HLP 代码 · 共 122 行

{smcl}
{* 14feb2005}{...}
{cmd:help centile} {right:dialog:  {bf:{dialog centile}}}
{hline}

{title:Title}

{p2colset 5 20 22 2}{...}
{p2col :{hi:[R] centile} {hline 2}}Report centile and confidence
interval{p_end}
{p2colreset}{...}


{title:Syntax}

{p 8 16 2}
{cmd:centile} [{varlist}] {ifin}
[{cmd:,}
{it:options}]

{synoptset 20 tabbed}{...}
{synopthdr}
{synoptline}
{syntab:Main}
{synopt :{opth c:entile(numlist)}}report specified centiles;
default is {cmd:centile(50)}{p_end}

{syntab:Options}
{synopt :{opt cc:i}}binomial exact; conservative confidence interval{p_end}
{synopt :{opt n:ormal}}normal, based on observed centiles{p_end}
{synopt :{opt m:eansd}}normal, based on mean and standard deviation{p_end}
{synopt :{opt l:evel(#)}}set confidence level; default is {cmd:level(95)}{p_end}
{synoptline}
{p2colreset}{...}
{p 4 6 2}
{opt by} may be used with {cmd:centile}; see {helpb by}.


{title:Description}

{pstd}
{cmd:centile} estimates specified centiles and calculates confidence
intervals. If no {varlist} is specified,
{cmd:centile} calculates centiles for all the variables in the dataset.
If {cmd:centile()} is not specified, medians ({cmd:centile(50)}) are reported.

{pstd}
By default, {cmd:centile} uses a binomial method for obtaining confidence
intervals that makes no assumptions as to the underlying distribution of the
variable.


{title:Options}

{dlgtab:Main}

{phang}
{opth centile(numlist)} specifies the centiles to be reported.
The default is to display the 50th centile.
Specifying
{cmd:centile(5)} requests that the 5th centile be reported.
Specifying {cmd:centile(5 50 95)} requests that the 5th, 50th, and 95th
centiles be reported.
Specifying {cmd:centile(10(10)90)} requests that the 10th, 20th, ..., 90th
centiles be reported; see {help numlist}.

{dlgtab:Options}

{phang}
{opt cci} (conservative confidence interval) forces the confidence limits to
fall exactly on sample values.  Confidence intervals displayed with the
{opt cci} option are slightly wider than those with the default ({opt nocci})
option.

{phang}
{opt normal} causes the confidence interval to be calculated using a formula
for the standard error of a normal-distribution quantile given by Kendall and
Stuart. The {opt normal} option is useful when you want empirical
centiles{hline 2}that is, centiles based on sample order statistics rather
than on the mean and standard deviation{hline 2}and are willing to assume
normality.

{phang}
{opt meansd} causes the centile and confidence interval to be calculated based
on the sample mean and standard deviation, and assumes normality.

{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}.


{title:Examples}

{phang}
{cmd:. centile x, centile(5 50 95)}{p_end}
{p 15 15 2}calculates the 5th, 50th (median), and 95th centiles for {cmd:x}, not
assuming normality.

{phang}
{cmd:. centile x y, centile(2.5 97.5) meansd}{p_end}
{p 15 15 2}calculates 2.5th and 97.5th centiles for {cmd:x} and {cmd:y} from
their means and standard deviations, assuming normality.

{phang}
{cmd:. centile x, centile(10) normal}{p_end}
{p 15 15 2}calculates the 10th centile for {cmd:x}, assuming normality.

{pstd}
In each case, 95% confidence intervals for the specified centiles are
provided.


{title:Also see}

    Manual:  {hi:[R] centile}

{psee}
Online:  {helpb ci}, {helpb pctile}, 
{helpb summarize}
{p_end}