random_RR: Random generation of log response ratio (RR) effect sizes.
Description
Generates random log response ratios and their variances (Hedges et al. 1999).
NOTE: samples from a log-normal distribution to generate non-negative control
and treatment means (following Lajeunesse 2015).
Usage
random_RR(K, X_t, var_t, N_t, X_c, var_c, N_c)
Arguments
K
Number of effect sizes to generate.
X_t
The population mean (mu) of the (t)reatment group (numerator of
ratio).
var_t
The population variance of the treatment group mean.
N_t
The number of samples of the treatment mean. When a non-negative
integer, all treatment means will be estimated using the same N. A
vector of unequal N's can also be taken; if so, K will be ignored and the
number of randomly generated means will equa
X_c
The population mean (mu) of the (c)ontrol group (denominator of
ratio).
var_c
The population variance of the control group mean.
N_c
The number of samples of the control mean. When a non-negative
integer, all control means will be estimated using the same N. A
vector of unequal N's can also be taken; if so, K will be ignored and the
number of randomly generated means will
Value
A data table with columns of random effect sizes (RR) and their
variances.
References
Hedges, L.V., J. Gurevitch, and P.S. Curtis. 1999. The
meta-analysis of response ratios in experimental ecology. Ecology 80:
1150-1156.
Lajeunesse, M.J. 2015. Bias and correction for the log response
ratio used in ecological meta-analysis. Ecology.