combine.test: Function to combine probabilities
In bhklab/survcomp: Performance Assessment and Comparison for Survival Analysis

Description Usage Arguments Details Value Author(s) References See Also Examples

The function combines several p-value estimated from the same null hypothesis in different studies involving independent data.

1 2	combine.test(p, weight, method = c("fisher", "z.transform", "logit"), hetero = FALSE, na.rm = FALSE)

`p`	vector of p-values
`weight`	vector of weights (e.g. sample size of each study)
`method`	`fisher` for the Fisher's combined probability test, `z.transform` for the Z-transformed test, `logit` for the weighted Z-method
`hetero`	`TRUE` is the heterogeneity should be taken into account, `FALSE` otherwise
`na.rm`	`TRUE` if the missing values should be removed from the data, `FALSE` otherwise

The p-values must be one-sided and computed from the same null hypothesis.

p-value

Benjamin Haibe-Kains

Whitlock, M. C. (2005) "Combining probability from independent tests: the weighted Z-method is superior to Fisher's approach", J. Evol. Biol., 18, pages 1368–1373.

test.hetero.test

p <- c(0.01, 0.13, 0.07, 0.2)
w <- c(100, 50, 200, 30)

#with equal weights
combine.test(p=p, method="z.transform")
#with p-values weighted by the sample size of the studies
combine.test(p=p, weight=w, method="z.transform")