betaPLS: Calculates penalized least squares estimates for orthogonal...

Description Usage Arguments Value Author(s)

View source: R/betaPLS.R

Description

Calculates penalized least squares estimates for orthogonal design matrix, based on ordinary least squares estimates.

Usage

1
2
betaPLS(coeftab, lambda, penalty.type = "L0", fnlambda = function(b1,
  b2) exp(pmax(b1, b2)/2), a = 3.7)

Arguments

lambda

a constant.

fnlambda

a function that maps each pair of frequencies to a constant between 0 and 1, corresponding to how the penalty relates to n and m.

df.beta

a data.frame representation of the OLS coefficients (result of buildFrequencyDf()).

type

the type of penalty. "L0" corresponds to hard thresholding, "L1" corresponds to LASSO or "SCAD".

Value

a data.frame with rows corresponding to combinations of frequencies and functions and their respective real-valued fourier transform coefficient.

Author(s)

Victor Freguglia Souza


VicFreguglia/GibbsRF documentation built on Oct. 25, 2019, 11:19 p.m.