preprocess {autoMFA}

Preprocess

Description

Performs the pre-processing of a data matrix such that it is ready to be used by vbmfa.

Usage

preprocess(Y, ppp, shrinkQ)

Arguments

Y

An n by p data matrix which is to be scaled.

ppp

An optional p by 2 matrix where the columns represent the sample mean and sample standard deviation of the pth dimension of Y.

shrinkQ

If 1, the data is shrunk according to ppp. If 0, the data is expanded to invert a prior shrinking by ppp.

Value

A list containing

• Yout: A processed data matrix of observations.

• ppp: The shrinkage which as applied in the processing.

References

Ghahramani Z, Beal MJ (2000). “Variational inference for Bayesian Mixtures of Factor Analysers.” In Advances in neural information processing systems, 449–455.

See Also

vbmfa for fitting models after using preprocess.

Examples

Yout <- preprocess(autoMFA::MFA_testdata);