Stata 15 help for brr_options

[SVY] brr_options -- More options for BRR variance estimation

Syntax

brr_options Description ------------------------------------------------------------------------- SE mse use MSE formula for variance nodots suppress replication dots dots(#) display dots every # replications hadamard(matrix) Hadamard matrix fay(#) Fay's adjustment

saving(filename, ...) save results to filename verbose display the full table legend noisily display any output from command trace trace command title(text) use text as the title for results nodrop do not drop observations reject(exp) identify invalid results ------------------------------------------------------------------------- saving(), verbose, noisily, trace, title(), nodrop, and reject() are not shown in the dialog boxes for estimation commands.

Description

svy accepts more options when performing BRR variance estimation. See [SVY] svy brr for a complete discussion.

Options

+----+ ----+ SE +---------------------------------------------------------------

mse specifies that svy compute the variance by using deviations of the replicates from the observed value of the statistics based on the entire dataset. By default, svy computes the variance by using the deviations of the replicates from their mean.

nodots suppresses display of the replication dots. By default, one dot character is printed for each successful replication. A red `x' is displayed if command returns with an error, and `e' is displayed if at least one of the values in exp_list is missing.

dots(#) displays dots every # replications. dots(0) is a synonym for nodots.

hadamard(matrix) specifies the Hadamard matrix to be used to determine which PSUs are chosen for each replicate.

fay(#) specifies Fay's adjustment. This option overrides the fay(#) option of svyset; see [SVY] svyset.

saving(), verbose, noisily, trace, title(), nodrop, reject(); see [SVY] svy brr.


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index