fitBR: learn a baseline regularization (BR) model

In sverchkov/BaselineRegularization: Baseline Regularization

Description

learn a BR model from matrix and vector input currently only support ridge penalty onver drug effects currently only support one pair of lambda1 and lambda2 learning at a time

Usage

fitBR(interval_baseline_parameter, baseline_parameter_obs_period, X, l, n,
  lambda1, lambda2, lambda3, ...)

Arguments

interval_baseline_parameter	An array where each position corresponds to an interval and each value corresponds to the index of a baseline parameter
baseline_parameter_obs_period	An array where each position corresponds to a baseline parameter and each value corresponds to the index of an observation period
X	data matirx, each row represent data from one interval, each column represent drug exposure info
l	length of each interval
n	number of condition occurrences on each interval
lambda1	strength of ridge regression for beta
lambda2	strengh of fused lasso for baseline
lambda3	strength of ridge regression for baseline
...	currently unused

Value

a br model

Author(s)

Zhaobin Kuang

Yuriy Sverchkov

sverchkov/BaselineRegularization documentation built on May 9, 2019, 1:26 p.m.