Learn R Programming
⚠️There's a newer version (1.2.17) of this package.Take me there.

CorReg (version 0.14.2)

Linear regression based on linear structure between covariates

Description

Sequential linear regression based on a recursive structural equation model (explicit correlations). It permits to face highly correlated datasets.

Copy Link

Version

Install

install.packages('CorReg')

Monthly Downloads

81

Version

0.14.2

License

CeCILL

Maintainer

Clement THERY

Last Published

February 14th, 2014

Functions in CorReg (0.14.2)

cleanZR2

To clean Z based on R2
compare_struct

To compare structures (Z)
CorReg-package

Algorithms for regression with correlated covariates
cleancolZ

clean Z columns (if BIC improved)
searchZ_sparse

Sparse structure research
density_estimation

BIC of estimated marginal gaussian mixture densities
searchZ

MCMC algo to find a structure between the covariates
Y_generator

Response variable generator with a linear model
ProbaZ

Probability of Z without knowing the dataset. It also gives the exact number of binary nilpotent matrices of size p.
WhoIs

Give the partition implied by a structure
cleanZ

clean Z (if BIC improved)
mixture_generator

Gaussiam mixture dataset generator with regression between the covariates
MSE_loc

simple MSE function
Winitial

initialization based on a wheight matrix (correlation or other)
MSEZ

Computes the MSE on the joint distribution of the dataset
readY

a summary-like function
compare_sign

compare signs of the coefficients in two vectors
correg

Estimates the response variable using a structure
BicZcurve

Curve of the BIC for each possible p2 with a fixed Z and truncature of Z
R2Z

Estimates R2 of each subression
readZ

read the structure and explain it
OLS

Ordinary Least Square efficiently computed with SEM for missing values
hatB

Estimates B matrix
compare_zero

compare 0 values in two vectors
BicZ

Compute the BIC of a given structure
cleanYtest

Selection method based on p-values (coefficients)