Fit a linear model via penalized nonconvex loss function.
# S3 method for formula
ncl(formula, data, weights, offset=NULL, contrasts=NULL,
x.keep=FALSE, y.keep=TRUE, ...)
# S3 method for matrix
ncl(x, y, weights, offset=NULL, ...)
# S3 method for default
ncl(x, ...)An object with S3 class "ncl" for the various types of models.
the call that produced this object
predicted values
pseudo response values in the MM algorithm
symbolic description of the model, see details.
argument controlling formula processing
via model.frame.
optional numeric vector of weights. If standardize=TRUE, weights are renormalized to weights/sum(weights). If standardize=FALSE, weights are kept as original input
input matrix, of dimension nobs x nvars; each row is an observation vector
response variable. Quantitative for rfamily="clossR" and -1/1 for classification.
Not implemented yet
the contrasts corresponding to levels from the
respective models
For glmreg: logical values indicating whether the response vector and model matrix used in the fitting process should be returned as components of the returned value. For ncl_fit: x is a design matrix of dimension n * p, and x is a vector of observations of length n.
Other arguments passing to ncl_fit
Zhu Wang <zwang145@uthsc.edu>
The robust linear model is fit by majorization-minimization along with linear regression. Note that the objective function is $$weights*loss$$.
Zhu Wang (2021), MM for Penalized Estimation, TEST, tools:::Rd_expr_doi("10.1007/s11749-021-00770-2")
#binomial
x=matrix(rnorm(100*20),100,20)
g2=sample(c(-1,1),100,replace=TRUE)
fit=ncl(x,g2,s=1,rfamily="closs")
Run the code above in your browser using DataLab