Estimates the variation matrix of count-compositional data
based on a the same approximation used in logitNormalVariation()
only for this function it uses empirical estimates of mu and Sigma.
Also performs zero-imputation using cmultRepl() from the zCompositions package.
Usage
pluginVariation(
counts,
type = c("standard", "phi", "phis", "rho"),
order = c("second", "first"),
impute.zeros = TRUE,
...
)
Value
An estimate of the requested metric of proportionality.
Arguments
counts
Matrix of counts; samples are rows and features are columns.
type
Type of variation metric to be calculated: standard, phi,
phis (a symmetrical version of phi), rho, or logp (the variance-covariance matrix of log-transformed proportions).
order
The order of the Taylor-series approximation to be used in the
estimation
impute.zeros
If TRUE, then cmultRepl() from the zCompositions package is used to impute zero values in the counts matrix.
...
Optional arguments passed to zero-imputation function cmultRepl()