meanz: Combine p values using the mean z method

In metap: Meta-Analysis of Significance Values

Combine p values using the mean z method

Description

Combines p values using the mean of z method\loadmathjax

Usage

meanz(p, log.p = FALSE)
## S3 method for class 'meanz'
print(x, ...)

Arguments

p	\sigvec
log.p	\logp
x	An object of class ‘meanz’
...	Other arguments to be passed through

Details

Let \mjdeqn\barz = \sum_i=1^k \fracz(p_i)kbarz = sum(z(p) / k) and \mjdeqns_\barz = \fracs_z\sqrtks_barz = s_z / sqrt k Defined as \mjdeqn \frac\barzs_\barz > t_k-1(\alpha) ((barz / s_barz) > t_k-1(alpha)

\lele

two As can be seen if all the \mjseqnp_i are equal or close to equal this gives a \mjeqnt=\pm\inftyt=+-infty leading to a returned value of 0 or 1. A set of \mjseqnp values with small variance will necessarily give a large value for \mjdeqn\frac\barzs_\barz((barz / s_barz) and hence a small \mjseqnp value which may be smaller than that for another set all of whose primary values are less than any in the first set. See examples for a demonstration.

\plotmethod

Value

An object of class ‘meanz’ and ‘metap’, a list with entries

z	The value of the mean \mjseqnz statistic
p	The associated \mjseqnp value
validp	The input vector with illegal values removed

Author(s)

Michael Dewey

References

\insertRef

becker94metap

See Also

See also plotp

Examples

data(dat.metap)
beckerp <- dat.metap$beckerp
meanz(beckerp)
meanz(c(0.1, 0.2))  # greater than next example
meanz(c(0.3, 0.31)) # less than above
all.equal(exp(meanz(beckerp, log.p = TRUE)$p), meanz(beckerp)$p)

metap documentation built on Sept. 11, 2024, 6:53 p.m.