{smcl}
{hline}
help for {hi:svrmean}, {hi:svrtotal}, {hi:svrratio}
{hline}
{title:Estimate means, totals, ratios, and proportions for survey data,}
{title:with replication-based standard errors}
{p 8 17}{cmd:svrmean}{space 2}{it:varlist} [{cmd:if} {it:exp}]
[{cmd:in} {it:range}] [{cmd:,} {it:common_options}]
{p 8 17}{cmd:svrtotal} {it:varlist} [{cmd:if} {it:exp}]
[{cmd:in} {it:range}] [{cmd:,} {it:common_options}]
{p 8 17}{cmd:svrratio} {it:varname} [{cmd:/}] {it:varname} [{it:varname}
[{cmd:/}] {it:varname ...}] [{cmd:if} {it:exp}]
[{cmd:in} {it:range}] [{cmd:,} {it:common_options}]
{p}where {it:common_options} are
{p 17 17}{cmd:by(}{it:varlist}{cmd:)}
[{cmdab:com:plete}|{cmdab:av:ailable}]
{cmdab:nolab:el}
{cmdab:l:evel(}{it:#}{cmd:)} {cmd:ci} {cmd:deff} {cmd:deft} {cmd:meff}
{cmd:meft} {cmd:obs} {cmd:size}
{p}{cmd:svrmean}, {cmd:svrratio}, and {cmd:svrtotal} typed without arguments
redisplay previous results. Any of the following options can be used when
redisplaying results:
{p 17 17}{cmdab:l:evel(}{it:#}{cmd:)} {cmd:ci} {cmd:deff} {cmd:deft}
{cmd:meff} {cmd:meft} {cmd:obs} {cmd:size}
{p}These commands are for use with replication weights. You must set your data for
replication based survey estimation with {help svrset} or {help survwgt} before
using these commands.
{title:Description}
{p}{cmd:svrmean}, {cmd:svrtotal}, and {cmd:svrratio} produce
estimates of population means, totals, ratios, and proportions.
Standard errors are calculated using a series of user-supplied replication weights, by
balanced repeated replication (BRR) or the survey jackknife (JK1, JK2, or JKn).
This is an alternate method to the Taylor series linearization methods
used by Stata's {help svy} commands. See {help survwgt} for details on the
creation of weights and estimation of variances with replication.
{p}Estimates for multiple subpopulations can be obtained using the {cmd:by()}
option. The {cmd:if} option will give estimates for a single population. (Note that
with replication methods, use of {cmd:if} or {cmd:in} produces correct estimates for
the relevant subpopulation; this is not the case with {help svy} commands.)
{p}{help svytest} will operate after these commands, as it does after their svy-based
equivalents.
{title:Options}
{p 0 4}{cmd:by(}{it:varlist}{cmd:)} specifies that estimates be computed for
the subpopulations defined by different values of the variable(s) in the
{it:varlist}.
{p 0 4}{cmd:nolabel} can only be specified when {cmd:by()} is specified.
{cmd:nolabel} requests that numeric values rather than value labels be used to
label output for subpopulations. By default, value labels are used.
{p 0 4}[{cmd:complete}|{cmd:available}] specifies how missing values
are to be handled. {cmd:complete} specifies that only observations with
complete data should be used. {cmd:available} specifies that all available
nonmissing values be used for each estimate.
{p 4 4}If neither {cmd:complete} nor {cmd:available} is specified,
{cmd:available} is the default when there are missing values and there are two
or more variables in the {it:varlist} (or four or more for {cmd:svyratio}).
{cmd:complete} must be specified to compute the covariance or to use
{cmd:svytest} after running the command; see help
{help svytest}.
{p 0 4}{cmd:level(}{it:#}{cmd:)} specifies the confidence level (i.e., nominal
coverage rate), in percent, for confidence intervals; see help {help level}.
{p 0 4}{cmd:ci} requests that confidence intervals be displayed. If no
display options are specified then, by default, confidence intervals are
displayed.
{p 0 4}{cmd:deff} requests that the design-effect measure deff be displayed.
If no display options are specified then, by default, deff is displayed.
{p 0 4}{cmd:deft} requests that the design-effect measure deft be displayed.
See {hi:[R] svymean} for a discussion on deff and deft.
{p 0 4}{cmd:meff} requests that the meff measure of misspecification effects
be displayed.
{p 0 4}{cmd:meft} requests that the meft measure of misspecification effects
be displayed. See {hi:[R] svymean} for a discussion of meff and meft.
{p 0 4}{cmd:obs} requests that the number of observations used for the
computation of the estimate be displayed for each row of estimates.
{p 0 4}{cmd:size} requests that the estimate of the (sub)population size be
displayed for each row of estimates. The (sub)population size estimate equals
the sum of the weights for those observations used for the mean/total/ratio
estimate.
{title:Examples}
{p 8 12}{inp:. svrmean birthwgt {p_end}
{p 8 12}{inp:. svrmean birthwgt, by(race)}{p_end}
{p 8 12}{inp:. svrmean birthwgt if race==1}
{p 8 12}{inp:. svrratio hdresult/tcresult}{p_end}
{title:Methods and formulae}
{p}See {help survwgt}.
{title:Acknowledgements}
{p}These commands consists largely of the ado file code from official Stata's {cmd:svy_x} command, which
underlies {cmd:svymean}, {cmd:svytotal}, and {cmd:svyratio}. They are
modified to calculate (co)variances differently.
I would like to thank Bobby Gutierrez at StataCorp for advice on implementation of BRR.
{title:Author}
Nick Winter
Cornell University
nw53@cornell.edu