md.vcov: Covariance matrix for mean differences

Description Usage Arguments Value Author(s) References Examples

Description

Compute variance-covariance matrix for multivariate meta-analysis when effect size is mean difference.

Usage

1
md.vcov(r,nt,nc,n_rt=0,n_rc=0,sdt,sdc)

Arguments

r

A list of correlation coefficient matrices of the outcomes from the studies. r[[k]][i,j] is the correlation coefficient between outcome i and outcome j from study k.

nt

A matrix with sample sizes in the treatment group reporting each of the outcome. nt[i,j] is the sample size from study i reporting the outcome j.

nc

Defined in a similar way as nt for control group.

n_rt

A list of matrices storing sample sizes in the treatment group reporting pairwised outcomes in the off diagonal elements. n_rt[[k]][i,j] is the sample size reporting both outcome i and outcome j from study k. Diagonal elements of these matrices are not used. The default value is zero, which means the smaller sample size reporting the corresponding two outcomes: i.e. n_rt[[k]][i,j]=min(nt[k,i],nt[k,j]).

n_rc

Defined in a similar way as n_rt for control group.

sdt

Sample standard deviation from each of the outcome. sdt[i,j] is the sample standard deviation from study i for outcome j.

sdc

Defined in a similar way as sdt for control group.

Value

list.md.cov A list of computed variance-covariance matrices.
md.cov A matrix whose rows are computed variance-covariance vectors.

Author(s)

Min Lu

References

Ahn, S., Lu, M., Lefevor, G.T., Fedewa, A. & Celimli, S. (2016). Application of meta-analysis in sport and exercise science. In N. Ntoumanis, & N. Myers (Eds.), An Introduction to Intermediate and Advanced Statistical Analyses for Sport and Exercise Scientists (pp.233-253). Hoboken, NJ: John Wiley and Sons, Ltd.

Wei, Y., & Higgins, J. (2013). Estimating within study covariances in multivariate meta-analysis with multiple outcomes. Statistics in Medicine, 32(7), 119-1205.

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
######################################################
# Example: Geeganage2010 data
# Preparing covariances for multivariate meta-analysis
######################################################
## set the correlation coefficients list r
r12=0.71
r.Gee=lapply(1:nrow(Geeganage2010),function(i){matrix(c(1,r12,r12,1),2,2)})

computvocv<-md.vcov(nt=subset(Geeganage2010, select=c(nt_SBP,nt_DBP)),
                    nc=subset(Geeganage2010, select=c(nc_SBP,nc_DBP)),
                    sdt=subset(Geeganage2010, select=c(sdt_SBP,sdt_DBP)),
                    sdc=subset(Geeganage2010, select=c(sdc_SBP,sdc_DBP)),
                    r=r.Gee)
# name variance-covariance matrix as covars
covars = computvocv$md.cov

#####################################################
# Running random-effects model using package "mvmeta"
#####################################################
#library(mvmeta)
#mvmeta_RE = summary(mvmeta(cbind(MD_SBP,MD_DBP),S=covars,
#                         data=subset(Geeganage2010,select=c(MD_SBP,MD_DBP)),
#                         method="reml"))
#mvmeta_RE

Example output

Loading required package: corpcor

metavcov documentation built on May 2, 2019, 4:15 a.m.