Learn R Programming

selectMeta (version 1.0.8)

selectMeta-package: Estimation of Weight Functions in Meta Analysis

Description

Publication bias, the fact that studies identified for inclusion in a meta analysis do not represent all studies on the topic of interest, is commonly recognized as a threat to the validity of the results of a meta analysis. One way to explicitly model publication bias is via selection models or weighted probability distributions. For details we refer to Iyengar & Greenhouse (1998), Dear & Begg (1992), and Rufibach (2011). In this package we provide implementations of all the weight functions proposed in these papers. The novelty in Rufibach (2011) is the proposal of a non-increasing variant of the nonparametric weight function of Dear & Begg (1992). Since virtually all parametric weight functions proposed so far in the literature are in fact decreasing and only few studies are included in a typical meta analysis regularization by imposing monotonicity seems a sensible approach. The new approach potentially offers more insight in the selection process than other methods, but is more flexible than parametric approaches. To maximize the log-likelihood function proposed by Dear & Begg (1992) under a monotonicity constraint on $w$ we use a differential evolution algorithm proposed by Ardia et al (2010a, b) and implemented in Mullen et al (2009).

The main functions in this package are IyenGreen and DearBegg. Using DearBeggMonotoneCItheta one can compute a profile likelihood confidence interval for the overall effect size $\theta$ and using DearBeggMonotonePvalSelection the simulation-based $p$-value to assess the null hypothesis of no selection, as described in Rufibach (2011, Section 6), can be computed. In addition, we provide two datasets: education, a dataset frequently used in illustration of meta analysis and passive_smoking, a second dataset that has caused some controversy about whether publication bias is present in this dataset or not.

Arguments

Details

Package:
selectMeta
Type:
Package
Version:
1.0.8
Date:
2015-07-03
License:
GPL (>=2)

References

Ardia, D., Boudt, K., Carl, P., Mullen, K.M., Peterson, B.G. (2010). Differential Evolution ('DEoptim') for Non-Convex Portfolio Optimization.

Ardia, D., Mullen, K.M., et.al. (2010). The 'DEoptim' Package: Differential Evolution Optimization in 'R'. Version 2.0-7.

Dear, K.B.G. and Begg, C.B. (1992). An Approach for Assessing Publication Bias Prior to Performing a Meta-Analysis. Statist. Sci., 7(2), 237--245.

Hedges, L. and Olkin, I. (1985). Statistical Methods for Meta-Analysis. Academic, Orlando, Florida.

Iyengar, S. and Greenhouse, J.B. (1988). Selection models and the file drawer problem. Statist. Sci., 3, 109--135.

Mullen, K.M., Ardia, D., Gil, D.L., Windover, D., Cline, J. (2009). 'DEoptim': An 'R' Package for Global Optimization by Differential Evolution.

Rufibach, K. (2011). Selection Models with Monotone Weight Functions in Meta-Analysis. Biom. J., 53(4), 689--704.

Examples

Run this code
# All functions in this package are illustrated 
# in the help file for the function DearBegg().

Run the code above in your browser using DataLab