lm.ols.refit: Get the ordinary least square estimated coefficients on a set...

In CSUV: Combined Selection and Uncertainty Visualiser (CSUV)

Description

Get the ordinary least square estimated coefficients on a set of previously selected covariates

Usage

lm.ols.refit(X, Y, intercept, est.betas, log.level = NULL)

Arguments

X	covariates (n times p matrix, n: number of entries, p: number of covariates)
Y	response (vector with n entries)
intercept	TRUE to fit the data with an intercept, FALSE to fit the data without an intercept
est.betas	estimated betas from previous fitted result. It can be a vector with p+1 entries (first entry as intercept) or a matrix with p+1 columns. Non-zero coefficient means the corresponding covariate is selected
log.level	log level to set. Default is NULL, which means no change in log level. See the function CSUV::set.log.level for more details

Value

a list of estimated coefficients

Examples

X = matrix(rnorm(1000), nrow = 100)
Y = rowSums(X[,1:3])+rnorm(100)
est.beta = rep(0, 11)
est.beta[2:5] = 1
ols.mod = lm.ols.refit(X, Y, intercept = FALSE, est.betas = est.beta)
print(ols.mod$est.b)

CSUV documentation built on Oct. 23, 2020, 5:49 p.m.