lasso.coef: Measure an impact of the covariates on the response using...
In rbvs: Ranking-Based Variable Selection

Description Usage Arguments Details Author(s) References

Measure an impact of the covariates on the response using Lasso This function evaluates the Lasso coefficients regressing y onto the design matrix x over subsamples in subsamples.

1
2
3

lasso.coef(x, y, subsamples, nonzero = NULL, family = c("gaussian",
  "binomial"), alpha = 1, maxit = 500, tol = 0.01, lambda.ratio = 1e-06,
  nlam = 25, ...)

`x`	Matrix with `n` observations of `p` covariates in each row.
`y`	Response vector with `n` observations.
`subsamples`	Matrix with `m` indices of `N` subsamples in each column.
`nonzero`	Number of non-zero coefficients estimated for each subsample.
`family`	Determines the likelihood optimised in the estimation procedure.
`alpha`	Scalar between 0 and 1 determining l2 penalty (see details).
`maxit`	Maximum number of itarations when computing the lasso coefficients.
`tol`	Scalar determining convergence of the lasso algorithm used.
`lambda.ratio`	Scalar being a fraction of 1. Used in the lasso algorithm
`nlam`	Number of penalty parameters used in the lasso algorithm.
`...`	Not in use.

To solve the Lasso problem, we implement the coordinate descent algorithm as in Breheny Jian (2011).

Rafal Baranowski, Patrick Breheny

Tibshirani, Robert. "Regression shrinkage and selection via the lasso." Journal of the Royal Statistical Society. Series B (Methodological) (1996): 267-288.

Breheny, Patrick, and Jian Huang. "Coordinate descent algorithms for nonconvex penalized regression, with applications to biological feature selection." The Annals of Applied Statistics 5.1 (2011): 232.

rbvs documentation built on May 2, 2019, 7:31 a.m.