csr: Elliott, Gargano and Timmermann (2013) COMPLETE SUBSET...

View source: R/csr.R

csrR Documentation

Elliott, Gargano and Timmermann (2013) COMPLETE SUBSET REGRESSIONS

Description

Incorporates techniques in section 3.4 This step aims to reduce the computation burden when K is large, but it is not feasible when K is too large using standard R functions For example, when K = 50, k = 25, the vector 1:choose(50,25) consumes 941832.4 Gb memory, which is an astronomical number Future worke: see Boot and Nibbering (2019) Random subspace method.

Usage

csr(y, X, k, C.upper = 5000, intercept = FALSE)

Arguments

y

response variable

X

Predictor matrix

k

subset size

C.upper

maximum number of subsets to be combined

intercept

A boolean: include an intercept term or not


zhan-gao/LasForecast documentation built on Sept. 18, 2024, 9:40 p.m.