QUT-package

<p>Selection of a threshold parameter \(\lambda\) for GLM-lasso and Square-root lasso. The method consists in considering a null
model, finding the theoretical distribution of the threshold parameter under
the null, and setting \(\lambda\) to an upper quantile of that distribution. Although
this strategy does not use the data to select \(\lambda\) but simply considers
the behavior under the null model, it provides a theoretically and computationally
sound selection.</p>

lasso

threshold

regularization

square-root lasso

zero-threshold

Thresholding based tests for null hypothesis of the form A beta =c, and the Quantile Universal Threshold (QUT) for lasso regularization of Generalized Linear Models (GLM) and square-root lasso to obtain a sparse model with a good compromise between high true positive rate and low false discovery rate. Giacobino et al. (2017) <doi:10.1214/17-EJS1366>. Sardy et al. (2017) <arXiv:1708.02908>.

Package:	QUT
Type:	Package
License:	GPL-2

Get 50% off unlimited learning

QUT-package: Quantile Universal Threshold

Description

Arguments

Details

References