aorsf (version 0.1.4)

orsf_scale_cph: Scale input data

Description

These functions are exported so that users may access internal routines that are used to scale inputs when orsf_control_cph is used.

Usage

orsf_scale_cph(x_mat, w_vec = NULL)
orsf_unscale_cph(x_mat)

Value

the scaled or unscaled x_mat.

Arguments

x_mat

(numeric matrix) a matrix with values to be scaled or unscaled. Note that orsf_unscale_cph will only accept x_mat inputs that have an attribute containing transform values, which are added automatically by orsf_scale_cph.

w_vec

(numeric vector) an optional vector of weights. If no weights are supplied (the default), all observations will be equally weighted. If supplied, w_vec must have length equal to nrow(x_mat).

Details

The data are transformed by first subtracting the mean and then multiplying by the scale. An inverse transform can be completed using orsf_unscale_cph or by dividing each column by the corresponding scale and then adding the mean.

The values of means and scales are stored in an attribute of the output returned by orsf_scale_cph (see examples)

Examples

x_mat <- as.matrix(pbc_orsf[, c('bili', 'age', 'protime')])

head(x_mat)

x_scaled <- orsf_scale_cph(x_mat)

head(x_scaled)

attributes(x_scaled) # note the transforms attribute

x_unscaled <- orsf_unscale_cph(x_scaled)

head(x_unscaled)

# numeric difference in x_mat and x_unscaled should be practically 0
max(abs(x_mat - x_unscaled))