Output-content options for metaparm
Syntax
options Description ------------------------------------------------------------------------- by(varlist) Variables specifying by-groups sumvar(varlist) Variables to be summed in output dataset dfcombine(combination_rule) Rule for combining degrees of freedom idnum(#) Value of numeric dataset ID variable nidnum(newvarname) Name of numeric dataset ID variable idstr(string) Value of string dataset ID variable nidstr(newvarname) Name of string dataset ID variable format(formatting_list) Display formats for variables in the output dataset -------------------------------------------------------------------------
where combination_rule is
satterthwaite | constant
and formatting_list is a list of form
varlist_1 format_1 ... varlist_n format_n
Description
These options are available for metaparm but not for parmcip. They control the contents of the output dataset (or resultsset) created by metaparm.
Options
by(varlist) specifies a list of existing by-variables in the input dataset. metaparm creates an output dataset with one observation in each by-group, or with one observation only if by() is not specified, and data on estimates, P-values, z- or t-statistics and confidence limits for the weighted sums of parameters within the by-group, or in the whole input dataset if by() is not specified. The weightings for the weighted sums are specified using the weight specification.
sumvar(varlist) specifies a list of variables in the input dataset to be included in the output dataset, with values equal to their unweighted sums in the input dataset (if by() is not specified) or to their unweighted sums within the by-group (if by() is specified). For instance, if the input dataset contains one observation per study to be entered into a meta-analysis, and contains a variable N specifying the number of subjects in the study, then the user can specify sumvar(N), and N will be present in the output dataset, where it will contain the total number of subjects in all the studies.
dfcombine(combination_rule) specifies a rule for combining the degrees of freedom of the input parameters to define the degrees of freedom for the output parameters, if the t-distribution is used to define confidence limits and P-values. If dfcombine(satterthwaite) is specified, then the formula of Satterthwaite (1946) is used. If dfcombine(constant) is specified, then metaparm checks that the degrees of freedom are constant (or constant within by-groups if by( varlist) is specified), and then sets the output degrees of freedom to the constant input degrees of freedom. dfcombine() is set to satterthwaite by default, but is ignored if the t-distribution is not used to define confidence limits and P-values. The option dfcombine(constant) is useful if the input parameters are uncorrelated parameters belonging to the same model estimation with pooled degrees of freedom, such as group means estimated using the regress command with group membership indicators as X-variables, using the noconst option, and the user uses metaparm to estimate contrasts of interest, such as differences or interactions. In these circumstances, using regress without the robust option and using dfcombine(constant) with metaparm gives confidence limits and P-values equivalent to those of the equal-variance t-test. By contrast, if the user estimates group means using a separate constant-only model for each group, and then uses metaparm with the option dfcombine(satterthwaite), then the confidence limits and P-values are equivalent to those of the unequal-variance t-test.
idnum(#) specifies an ID number for the output dataset. It is used to create a numeric variable, with default name idnum, in the output dataset, with that value for all observations. This is useful if the output resultsset is concatenated with other resultssets using append, or using dsconcat if installed.
nidnum(newvarname) specifies a name for the numeric ID variable evaluated by idnum(). If idnum() is present and nidnum() is absent, then the name of the numeric ID variable is set to idnum.
idstr(string) specifies an ID string for the output dataset. It is used to create a string variable, with default name idstr, in the output dataset, with that value for all observations. This is useful if the output resultsset is concatenated with other resultssets using append, or using dsconcat if installed.
nidstr(newvarname) specifies a name for the string ID variable evaluated by idstr(). If idstr() is present and nidstr() is absent, then the name of the string ID variable is set to idstr.
format(varlist_1 format_1 ... varlist_n format_n) specifies a list of pairs of variable lists and display formats. The formats will be allocated to the variables in the output dataset specified by the corresponding varlists.
Author
Roger Newson, Imperial College London, UK. Email: r.newson@imperial.ac.uk
References
Satterthwaite, F. E. 1946. An approximate distribution of estimates of variance components. Biometrics Bulletin 2(6): 110-114.
Also see
Manual: [D] append, [D] format On-line: help for append, format help for parmest, parmby, parmcip, metaparm, metaparm_outdest_opts, parmcip_opts, metaparm_resultssets help for dsconcat if installed