Options for metaparm and parmcip
Syntax
options Description ------------------------------------------------------------------------- notdist Use Normal or t-distribution eform Estimates and confidence limits exponentiated float Numeric output variables of type float or less fast Calculate confidence limits without precautions estimate(varname) Name of input estimate variable stderr(varname) Name of input standard error variable dof(varname) Name of input degrees of freedom variable zstat(newvarname) Name of output z-statistic variable tstat(newvarname) Name of output t-statistic variable pvalue(newvarname) Name of output P-value variable stars(numlist) List of P-value thresholds for stars nstars(newvarname) Name of output stars variable level(numlist) Confidence level(s) for calculating confidence limits clnumber(numbering_rule) Numbering rule for naming confidence limit variables minprefix(prefix) Prefix for lower confidence limits maxprefix(prefix) Prefix for upper confidence limits replace Replace variables with same names as output variables -------------------------------------------------------------------------
where numbering_rule is
level | rank
Description
These options are available for metaparm and for parmcip. They control the calculation of confidence limits and P-values. They are not often used, as metaparm and parmcip have defaults for these options, which are usually sensible.
Options notdist specifies whether or not a t-distribution is used to calculate confidence limits. If tdist is specified, then a t-distribution is used. If notdist is specified, then a standard Normal distribution is used. If neither tdist nor notdist is specified by the user, then the option is set to tdist if the dof() option is set to the name of an existing variable, and is set to notdist otherwise. If tdist is specified with metaparm, then the degrees of freedom in the output dataset are calculated from the degrees of freedom and standard errors in the input dataset, using the Satterthwaite formula (Satterthwaite, 1946).
eform indicates that the input estimates are exponentiated, and that the input standard errors are multiplied by the exponentiated estimate, and that the output confidence limits are to be exponentiated. If eform is used with metaparm, then the estimate variable in the output dataset is exponentiated, and the standard error variable in the output dataset is multiplied by the exponentiated estimate variable.
float specifies that the numeric output variables will be created as type float or below. If float is unset, then the numeric output variables are created as type double. Note that all generated variables are compressed as much as possible without loss of information, whether or not float is specified.
fast is an option for programmers, and specifies that no action will be taken to restore the original data if the user presses Break. If used with metaparm, the fast option implies the norestore option (see metaparm_outdest_opts).
estimate(varname) specifies the name of the input variable containing estimates. It is set to estimate if not specified.
stderr(varname) specifies the name of the input variable containing standard errors. It is set to stderr if not specified.
dof(varname) specifies the name of the input variable containing degrees of freedom. It is set to dof if not specified.
zstat(newvarname) specifies the name of the output variable containing the z-statistics. It is set to z if not specified.
tstat(newvarname) specifies the name of the output variable containing the t-statistics. It is set to t if not specified.
pvalue(newvarname) specifies the name of the output variable containing the P-values. It is set to p if not specified.
stars(numlist) is used to generate a string variable with default name stars, containing the stars for the P-values. It works in the same way as the stars() option of parmest.
nstars(newvarname) specifies the name of the output variable containing the stars, if stars() is specified. If nstars() is not specified, then the name is set to stars.
level(numlist) specifies the confidence levels, in percent, for the confidence limit variables created in the output dataset. It works in the same way as the level() option of parmest.
clnumber(numbering_rule) specifies the rule used to number the names of the confidence limit variables created in the output dataset. It works similarly to the clnumber() option of parmest. However, with parmcip and metaparm, the user may specify prefixes other than min and max for the confidence limits, using the minprefix() and maxprefix() options.
minprefix(prefix) specifies the prefix used for naming the lower confidence limit variables. It is set to min if not specified. For instance, if the user specifies minprefix(inf), then the lower 95% confidence limit variable will be named inf95.
maxprefix(prefix) specifies the prefix used for naming the upper confidence limit variables. It is set to max if not specified. For instance, if the user specifies maxprefix(sup), then the upper 95% confidence limit variable will be named sup95.
replace is a parmcip option, ignored by metaparm. It specifies that, if there are existing variables in the input dataset with the same names as the generated variables specified by the zstat(), tstat(), pvalue(), clnumber(), minprefix() or maxprefix() options, then those variables will be replaced. Whether or not replace is specified, the names of the output variables are not allowed to clash with each other, with the input variables, or with the variables in the by() and sumvar() options of metaparm.
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: [U] 20 Estimation and postestimation commands On-line: help for estcom help for parmest, parmby, parmcip, metaparm, metaparm_outdest_opts, metaparm_content_opts, metaparm_resultssets, parmest_resultssets