rowZs: rowZs: Fast Calculation of Z-Scores by Row
In mwrowe/microRutils: Mike Rowe's Miscellaneous R Utility Functions

Description Usage Arguments Details Value Author(s) See Also

rowZs() calculates Z-scores across rows of a matrix using fast row methods.

1	rowZs(X, vs, robust = F, parts = F, na.rm = T, min.sd)

`X`	Numeric matrix, or object that can be coerced to matrix.
`vs`	Indices (numeric or character) of the columns of X to be used in the calculation of the mean or median and sd or mad of each row. If omitted, defaults to all columns of X.
`robust`	Logical. If TRUE, calculate Z-scores as (x-median)/mad; otherwise, as (x - mean) / sd.
`parts`	Logical; if TRUE, return the mean/median and sd/`mad` of each row as named attributes, CALCULATED ACROSS THE COLUMNS SPECIFIED BY vs.
`na.rm`	Logical; should NA's be omitted?
`min.sd`	Optional numeric value, quantile of the non-zero sd's (or mad's) to replace very low (~zero) sd or mad values with. A typical value might be in the range 0 to 0.1. Meant to prevent very large or unrealistic Z-scores when the sd is small by chance or because of pile-up at a signal limit (e.g., saturation).

This function is meant to help find samples or groups of samples with extreme values, typically relative to a control group (vs)

Returns a matrix with the same dimensions as X with z-scores for each element of X calculated from the row means/medians and sds/mads. If parts==TRUE, will have named attributes: "vs.mean" and "vs.sd" or "vs.median" and "vs.mad", plus "vs".

M.W.Rowe, mwr.stats@gmail.com

mad.

mwrowe/microRutils documentation built on June 12, 2021, 2:41 p.m.