# olasso: Error standard deviation estimation using organic lasso In natural: Estimating the Error Variance in a High-Dimensional Linear Model

## Description

Solve the organic lasso problem \tilde{σ}^2_{λ} = \min_{β} ||y - X β||_2^2 / n + 2 λ ||β||_1^2 with two pre-specified values of tuning parameter: λ_1 = log p / n, and λ_2, which is a Monte-Carlo estimate of ||X^T e||_∞^2 / n^2, where e is n-dimensional standard normal.

## Usage

 1 olasso(x, y, intercept = TRUE, thresh = 1e-08)

## Arguments

 x An n by p design matrix. Each row is an observation of p features. y A response vector of size n. intercept Indicator of whether intercept should be fitted. Default to be TRUE. thresh Threshold value for underlying optimization algorithm to claim convergence. Default to be 1e-8.

## Value

A list object containing:

n and p:

The dimension of the problem.

lam_1, lam_2:

log(p) / n, and an Monte-Carlo estimate of ||X^T e||_∞^2 / n^2, where e is n-dimensional standard normal.

a0_1, a0_2:

Estimate of intercept, corresponding to lam_1 and lam_2.

beta_1, beta_2:

Organic lasso estimate of regression coefficients, corresponding to lam_1 and lam_2.

sig_obj_1, sig_obj_2:

Organic lasso estimate of the error standard deviation, corresponding to lam_1 and lam_2.

## Examples

 1 2 3 set.seed(123) sim <- make_sparse_model(n = 50, p = 200, alpha = 0.6, rho = 0.6, snr = 2, nsim = 1) ol <- olasso(x = sim$x, y = sim$y[, 1])

natural documentation built on May 2, 2019, 1:05 p.m.