summary.hdlm: summarize coefficients from a "glmnet" object

In hdlm: Fitting High Dimensional Linear Models

Description

'summary' method for class '"lm"'.

Usage

## S3 method for class 'hdlm'
summary(object, ..., level=NULL)

Arguments

object	an object of class '"hdlm"', usually, a result of a call to 'hdlm'.
level	Determines which coefficients to print on regression table. Level = 1 gives only those with non-zero coefficents if pval.method is equal to 'mean' and only those with p-value < 1 otherwise. Level = 2 gives anything with non-zero coefficient or non-one p-value, and Level = 3 (or any other choice) gives all coefficients.
...	further arguments passed to or from other methods.

Details

When fitting a large model, it can be cumbersome to look at results for all of the variables; when there exists high correlation between variables, level = 2 is often preferable. The different behavior for pval.method equal to 'mean' is due to the fact that the mean method (or Bayes method, which sets pval.method to 'mean') gives many p-values which are close to, but not exactly equal to, one.

hdlm documentation built on May 2, 2019, 8:35 a.m.