-------------------------------------------------------------------------------help parmhet_basic_opts(Roger Newson) -------------------------------------------------------------------------------

Basic options forparmhetandparmiv

optionsDescription -------------------------------------------------------------------------by(varlist)Variables specifying by-groupseformEstimates and confidence limits exponentiatedfloatNumeric output variables of typefloator lessdfcombine(combination_rule)Rule for combining degrees of freedomivweight(newvarname)Name of generated inverse-variance weight variablesweight(newvarname)Name of generated semi-weight variablesstderr(newvarname)Name of generated semi-weight-based standard error variable -------------------------------------------------------------------------where

combination_ruleis

welch|constant

DescriptionThese options are used in a similar way by

parmhetandparmiv.

Options

by(varlist)specifies a list of existing by-variables in the input dataset. The heterogeneity-test statistics are computed within by-groups ifby()is specified. Ifby()is not specified, then the test statistics are computed for the whole input dataset. The output dataset (or resultsset) created byparmhethas one observation per by-group, ifby()is specified, and only one observation otherwise.

eformindicates that the input estimates are exponentiated, and that the input standard errors are multiplied by the exponentiated estimate.eformis used when the user chooses to calculate heterogeneity-test statistics for the logs of the input parameter estimates, as is usual when the input parameters are odds ratios, risk ratios, or geometric mean ratios. Otherwise, heterogeneity-test statistics are calculated for the input parameters themselves.

floatspecifies that the numeric output variables will be created as typefloator below. Iffloatis unset, then the numeric output variables are created as typedouble. Note that all generated variables are compressed as much as possible without loss of information, whether or notfloatis specified.

dfcombine(combination_rule)specifies a rule for combining the degrees of freedom of the input parameters to define the denominator degrees of freedom for theF-test statistic, if a degrees of freedom variable is specified by the user. Ifdfcombine(welch)is specified, thenparmhetandparmivuse the formula of Welch (1951), popularized by Cochrane (1954), to calculate theF-statistic and its denominator degrees of freedom. Ifdfcombine(constant)is specified, thenparmhetandparmivcheck that the input degrees of freedom are constant (or constant within by-groups ifby(varlist)is specified), and then sets the denominator degrees of freedom to the constant input degrees of freedom, and calculates the heterogeneityF-statistic by dividing the heterogeneity chi-squared statistic by the heterogeneity degrees of freedom.dfcombine()is set towelchby default, but is ignored if a degrees of freedom variable is not specified. The optiondfcombine(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 theregresscommand with group membership indicators asX-variables, using thenoconstoption, and the user usesparmhetorparmivto test for heterogeneity between groups. In these circumstances, usingregresswithout therobustoption and usingdfcombine(constant)withparmhetorparmivgivesP-values equivalent to those of the equal-varianceF-test.

ivweight(newvarname)specifies the name of an output variable, to be generated in the existing input dataset, containing inverse-variance weights for the corresponding parameter estimates. These inverse-variance weights can then be input as aweights to themetaparm, module of theparmestpackage, using the estimates, standard errors and degrees of freedom input toparmhetorparmiv, to output estimates, confidence intervals andP-values for summary parameters generated by a fixed-effect meta-analysis.

sweight(newvarname)specifies the name of an output variable, to be generated in the existing input dataset, containing semi-weights for the corresponding parameter estimates, as described in Cochrane (1954). These semi-weights can then be input as aweights to themetaparmmodule of theparmestpackage, using the estimates and degrees of freedom input toparmhetorparmivtogether with the standard errors generated using thesstderr()option, to output estimates, confidence intervals andP-values for summary parameters generated by a DerSimonian-Laird randomly-variable-effect meta-analysis, as defined by DerSimonian and Laird (1986).

sstderr(newvarname)specifies the name of an output variable, to be generated in the existing input dataset, containing semi-weight-based standard errors for the corresponding parameter estimates. These standard errors are equal to the inverse square roots of the semi-weights generated by thesweight()option. If calculated, they can be input as thestderr()option to themetaparmmodule of theparmestpackage, using the estimates and degrees of freedom input toparmhetorparmiv, together with aweights generated by thesweight()option, to output estimates, confidence intervals andP-values for summary parameters generated by a DerSimonian-Laird randomly-variable-effect meta-analysis, as defined by DerSimonian and Laird (1986).

AuthorRoger Newson, National Heart and Lung Institute, Imperial College London, UK. Email: r.newson@imperial.ac.uk

ReferencesCochrane, W. G. 1954. The combination of estimates from different experiments.

Biometrics10(1): 101-129.DerSimonian, R. and Laird, N. 1986. Meta-analysis in clinical trials.

Controlled Clinical Trials7(3): 177-188.Welch, B. L. 1951. On the comparison of several mean values: an alternative approach.

Biometrika36(3/4): 330-336.

Also seeManual:

[R] meta,[R] testOn-line: help forparmhet,parmiv,parmhet_resultsset_opts,parmhet_hettest_opts,parmhet_resultssethelp fortesthelp forparmest,parmby,parmcip,metaparm,metanif installed