Description Usage Arguments Details Value Author(s) References Examples
This function summarizes multiple estimates for a single parameter by assuming a fixed (i.e. common) effect or random effects across studies. The summary estimate is obtained by calculating a weighted mean that accounts for sample size and (in case random effects are assumed) for between-study heterogeneity.
1 2 3 4 5 6 7 8 9 10 11 12 13 |
r |
Vector of numerics containing the effect size of each study |
r.se |
Vector of numerics containing the standard error of the effect sizes |
r.vi |
Vector of numerics containing the sampling variance of the effect sizes |
method |
Character string specifying whether a fixed-effect or a random-effects model should be fitted.
A fixed-effect model is fitted when using |
test |
Optional character string when |
labels |
Optional vector of characters containing the labels for the studies |
na.action |
A function which indicates what should happen when the data contain NAs.
Defaults to |
n.chains |
Optional numeric specifying the number of chains to use in the Gibbs sampler ( |
pars |
Optional list with additional arguments. The width of confidence, credibility and prediction intervals is
defined by |
verbose |
If TRUE then messages generated during the fitting process will be displayed. |
... |
Additional arguments that are passed to rma or runjags (if |
Unless specified otherwise, all meta-analysis models assume random effects and are fitted using restricted
maximum likelihood estimation with the metafor package (Viechtbauer 2010). Further, confidence intervals for
the average performance are based on the Hartung-Knapp-Sidik-Jonkman method, to better account for the uncertainty
in the estimated between-study heterogeneity (Debray 2016). A Bayesian meta-analysis can be performed by specifying
method="BAYES"
. In that case, the R packages runjags and rjags must be installed.]
For random-effects models, a prediction interval for the pooled effect size is displayed. This interval predicts in what
range future effect sizes will fall given what has already been observed (Higgins 2009, Riley 2011).
For Bayesian meta-analysis models that involve the Gibbs sampler (method="BAYES"
), the R packages runjags
and rjags
must be installed. The Bayesian approach uses an uninformative Normal prior for the mean and a
uniform prior for the between-study variance of the pooled effect size (Higgins 2009). By default, the Normal prior
has a mean of 0 and a variance of 1000. These hyperparameters can, however, be altered through the
variables hp.mu.mean
and hp.mu.var
in the argument pars
. The prior distribution of the between-study
standard deviation is given by a uniform distribution, by default bounded between 0 and 100.
An object of the class uvmeta
for which many standard methods are available.
array with (transformed) data used for meta-analysis, and method(s) used for restoring missing information.
character string specifying the meta-analysis method.
estimated performance statistic of the model. For Bayesian meta-analysis, the posterior median is returned.
standard error (or posterior standard deviation) of the summary estimate.
estimated amount of (residual) heterogeneity. Always 0 when method="FE"
. For Bayesian meta-analysis, the posterior median is returned.
estimated standard error (or posterior standard deviation) of the between-study variation.
lower bound of the confidence (or credibility) interval of the summary estimate
upper bound of the confidence (or credibility) interval of the summary estimate
lower bound of the (approximate) prediction interval of the summary estimate
upper bound of the (approximate) prediction interval of the summary estimate
the full results from the fitted model
vector specifying the label of each study.
Thomas Debray <thomas.debray@gmail.com>
Biggerstaff BJ, Tweedie RL. Incorporating variability in estimates of heterogeneity in the random effects model in meta-analysis. Statistics in Medicine 1997; 16: 753–768.
Borenstein M, Hedges LV, Higgins JPT, Rothstein HR. A basic introduction to fixed-effect and random-effects models for meta-analysis. Research Synthesis Methods 2010; 1: 97–111. doi: 10.1002/jrsm.12
DerSimonian R, Laird N. Meta-analysis in clinical trials. Controlled Clinical Trials 1986; 7: 177–188.
Graham PL, Moran JL. Robust meta-analytic conclusions mandate the provision of prediction intervals in meta-analysis summaries. Journal of Clinical Epidemiology 2012; 65: 503–510.
Higgins JPT, Thompson SG. Quantifying heterogeneity in a meta-analysis. Statistics in Medicine 2002; 21: 1539–1558.
Higgins JPT, Thompson SG, Spiegelhalter DJ. A re-evaluation of random-effects meta-analysis. J R Stat Soc Ser A Stat Soc. 2009;172:137–59. doi: 10.1111/j.1467-985X.2008.00552.x
Riley RD, Higgins JPT, Deeks JJ. Interpretation of random effects meta-analyses. British Medical Journal 2011; 342: d549. doi: 10.1136/bmj.d549
Viechtbauer W. Conducting Meta-Analyses in R with the metafor Package. Journal of Statistical Software. 2010; 36. doi: 10.18637/jss.v036.i03
1 2 3 4 5 6 7 8 9 10 11 12 13 14 | data(Roberts)
# Frequentist random-effects meta-analysis
fit1 <- with(Roberts, uvmeta(r=SDM, r.se=SE, labels=rownames(Roberts)))
summary(fit1)
plot(fit1) #show a forest plot
fit1
## Not run:
# Bayesian random effects meta-analysis
fit2 <- with(Roberts, uvmeta(r=SDM, r.se=SE, labels=rownames(Roberts), method="BAYES"))
plot(fit2)
## End(Not run)
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.