This function unwinds the initial standardization of the data to obtain coefficient values on their original scale. It is called by plmm_format().
untransform(
std_scale_beta,
p,
std_X_details,
fbm_flag,
plink_flag,
use_names = TRUE
)a matrix of estimated coeffcients, 'beta_vals', that is on the scale of the original data.
The estimated coefficients on the standardized scale
The number of columns in the original design matrix
A list with 3 elements describing the standardized design matrix BEFORE rotation; this should have elements 'scale', 'center', and 'ns'
Logical: is the corresponding design matrix filebacked?
Logical: did these data come from PLINK files?
Note: This flag matters because of how non-genomic features
are handled for PLINK files -- in data from PLINK files,
unpenalized columns are not counted in the p argument. For delimited
files, p does include unpenalized columns. This difference has
implications for how the untransform() function determines the
appropriate dimensions for the estimated coefficient matrix it returns.
Logical: should names be added? Defaults to TRUE. Set to FALSE inside of cvf() helper, as 'ns' will vary within CV folds.