Variance estimation using refitted cross-validation in ultrahigh dimensional regression.
sigmarcv(y, X, cv = FALSE, fit = NA, intercept = TRUE)response variable. Quantitative for family=gaussian, or family=poisson (non-negative counts). For family=binomial should be a factor with two levels.
input matrix, of dimension n x p; each row is an observation vector.
when FALSE, variance is estimated using Refitted Cross Validation in Fan et al. 2012; and when TRUE, it is estimated using cross validation as in Reid et al. 2013. Default is FALSE.
A user supplied glmnet or lars object. Typical usage is to leave it empty so that the program computes the regularization path using the algorithm selected in type. WARNING: use with care, if supplied, object options must match with user supplied options.
should intercept(s) be fitted (default=TRUE) or set to zero (FALSE).
Estimator of \(\sigma\)
Jianqing Fan, Shaojun Guo and Ning Hao. Variance estimation using refitted cross-validation in ultrahigh dimensional regression. Journal of the Royal Statistical Society: Series B. 2012