{smcl} {hline} help for {hi:svrest} {hline} {title:Replication-based estimates of covariances for complex survey data} {p 4 23}{cmd:svrest}{space 6}{cmd:"}{it:command}{cmd:"} {cmd:"}{it:exp_list}{cmd:"} [{cmd:if} {it:exp}] [{cmd:in} {it:range}] [{cmd:,} [{cmdab:mat:rices(}{it:list}{cmd:)} {cmdab:noi:sily} {cmdab:or} {cmdab:li:ststats} {bind:{cmdab:diparm(}{it:eq type name} [{it:ci}] [ \ ...]{cmd:)}} ] {title:Description} {p}{cmd:svrest} runs the user-specified {it:command}, and calculates covariance estimates for the statistics specified in {it:exp_list}, using replication methods for complex survey data. {cmd:svrest} is based on {help bs}. {p}The expressions in {it:exp_list} must be separated by spaces and there must be no spaces within each expression. Note that {it:command} and {it:exp_list} must both be enclosed in double quotes. {p}See {help survwgt} and {help svr} for details on these replication methods. {p}{cmd:svrest} estimates the statistics jointly, so it calculates both the variances of each statistic, and the covariances among them. The results are stored as {help est:estimates}. {p}See {help svresttest} for testing hypotheses involving the estimated statistics generated by this command. {title:Options} {p 0 4}{cmd:matrices} indicates one entire matrices of results to be included. This allows the easy display of the entire coefficient vector when using {cmd:svrest} for an estimation command not supported by {help brrmodel}. For example, the following would compute replication-based standard errors for the full coefficient vector from a heckman model: {p 8 8}{inp:. svrest "heckman income age race weight, select(education)" , mat(e(b))} {p}{bf:Important note:} It is vital that the estimation in each replicate produce the same coefficient vector. Thus, for example, when running {cmd:mlogit}, be sure to include the {cmd:base()} option, so that the same category is omitted from the estimation for every replicate. {p 0 4}{cmd:dots} requests a dot be placed on the screen at the beginning of each replication, thus providing entertainment of a sort when long, slow commands are run. {p 0 4}{cmd:noisily} requests that output from each replication be displayed. {p 0 4}{cmd:or} requests that statistics be displayed as odds ratios. {p 0 4}{cmd:liststats} requests that a key be displayed, indicating what each statistic is. {p 0 4}{cmd:diparm} requests that single equation statistics be redisplayed in the natural metric. You must indicate the name of the equation containing the statistic, the type of transformation, the name for the statistic, and, optionally, {cmd:ci} to indicate that the confidence interval should be appropriately transformed. When specified, this option simply calls {help _diparm} to redisplay statistics that are not estimated in their natural metric. For example, for {cmd:heckman}, you might do the follwowing: {p 8 8}{inp:. svrest "heckman income age, select(educ)" , mat(e(b)) diparm(athrho exp rho ci \ lnsigma exp sigma di)} {title:Examples} {p 8 12}{inp:. svrest "reg mpg weight foreign" "_b[weight] _b[foreign]"} {p 8 12}{inp:. svrest "kappa cat1-cat3" "r(z)"} {title:Acknowledgements} {p}I would like to thank members of the Stata technical team, whose suggestions led to the creation of this command. {title:Author} Nick Winter Cornell University nw53@cornell.edu