| Title: | Robust Inference for Meta-Analysis with Influential Outlying Studies |
|---|---|
| Description: | Robust inference methods for fixed-effect and random-effects models of meta-analysis are implementable. The robust methods are developed using the density power divergence that is a robust estimating criterion developed in machine learning theory, and can effectively circumvent biases and misleading results caused by influential outliers. The density power divergence is originally introduced by Basu et al. (1998) <doi:10.1093/biomet/85.3.549>, and the meta-analysis methods are developed by Noma et al. (2022) <forthcoming>. |
| Authors: | Hisashi Noma [aut, cre], Shonosuke Sugasawa [aut], Toshi A. Furukawa [aut] |
| Maintainer: | Hisashi Noma <[email protected]> |
| License: | GPL-3 |
| Version: | 1.2-1 |
| Built: | 2025-01-31 02:52:48 UTC |
| Source: | https://github.com/cran/robustmeta |
A R package for implementing the robust inference methods for meta-analysis involving influential outlying studies.
Noma, H., Sugasawa, S. and Furukawa, T. A. (2022). Robust inference methods for meta-analysis involving influential outlying studies. In Preparation.
ID: Study ID
Souce: First author name and year of publication
m1: Estimated mean in experimental group
s1: Standard deviation in experimental group
n1: Number of observations in experimental group
m2: Estimated mean in control group
s2: Standard deviation in control group
n2: Number of observations in control group
data(clbp)data(clbp)
A data frame with 23 rows and 8 variables
Rubinstein, S. M,, de Zoete, A., van Middelkoop, M., Assendelft, W. J. J., de Boer, M. R., van Tulder, M. W. (2019). Benefits and harms of spinal manipulative therapy for the treatment of chronic low back pain: systematic review and meta-analysis of randomised controlled trials. BMJ. 364: l689.
Implementing the robust inference for meta-analysis involving influential outlying studies based on the density power divergence.
rmeta(y, v, model="RE", gamma=0.01)rmeta(y, v, model="RE", gamma=0.01)
y |
A vector of the outcome measure estimates (e.g., MD, SMD, log OR, log RR, log HR, RD) |
v |
A vector of the variance estimate of |
model |
Type of the pooling model; |
gamma |
Unit of grid search to explore the optimal value of tuning parameter |
Results of the robust inference for meta-analysis.
mu: Estimate of the common effect (for the fixed-effect model) or the grand mean (for the random-effects model).
se: Standard error estimate of mu.
CI: 95 percent confidence interval of mu.
P: P-value of the hypothesis test of mu=0.
alpha: Selected alpha by the Hyvarinen score.
W: Contribution rates of individual studies (ui: contribution rates of the conventional methods, wi: contribution rates of the robust methods).
Noma, H., Sugasawa, S. and Furukawa, T. A. (2022). Robust inference methods for meta-analysis involving influential outlying studies. In Preparation.
Basu, A., Harris, I. R., Hjort, N. L., Jones, M. C. (1998). Robust and efficient estimation by minimizing a density power divergence. Biometrika. 85: 549-559.
Sugasawa, S. and Yonekura, S. (2021). On selection criteria for the tuning parameter in robust divergence. Entropy. 23: 1147.
require(metafor) data(clbp) edat1 <- escalc(measure="SMD",m1i=m1,m2i=m2,sd1i=s1,sd2i=s2,n1i=n1,n2i=n2,data=clbp) DL1 <- rma(yi, vi, data=edat1, method="DL") print(DL1) # ordinary DerSimonian-Laird method plot(DL1) # plots of influential statistics, etc. ### y <- as.numeric(edat1$yi) # definition of summary statistics v <- edat1$vi rmeta(y,v) # robust inference based on the random-effects model rmeta(y,v,model="FE") # robust inference based on the fixed-effect modelrequire(metafor) data(clbp) edat1 <- escalc(measure="SMD",m1i=m1,m2i=m2,sd1i=s1,sd2i=s2,n1i=n1,n2i=n2,data=clbp) DL1 <- rma(yi, vi, data=edat1, method="DL") print(DL1) # ordinary DerSimonian-Laird method plot(DL1) # plots of influential statistics, etc. ### y <- as.numeric(edat1$yi) # definition of summary statistics v <- edat1$vi rmeta(y,v) # robust inference based on the random-effects model rmeta(y,v,model="FE") # robust inference based on the fixed-effect model
study: Study ID
d1: Number of depression events in treatment group
n1: Number of observations in treatment group
d0: Number of depression events in control group
n0: Number of observations in control group
data(varenicline)data(varenicline)
A data frame with 29 rows and 5 variables
Thomas, K. H., Martin, R. M., Knipe, D. W., Higgins, J. P., Gunnell, D. (2015). Risk of neuropsychiatric adverse events associated with varenicline: systematic review and meta-analysis. BMJ. 350: h1109.