Title: | Standard Error Adjusted Adaptive Lasso |
---|---|
Description: | Standard error adjusted adaptive lasso (SEA-lasso) is a version of the adaptive lasso, which incorporates OLS standard error to the L1 penalty weight. This method is intended for variable selection under linear regression settings (n > p). This new weight assignment strategy is especially useful when the collinearity of the design matrix is a concern. |
Authors: | Wei Qian <[email protected]> |
Maintainer: | Wei Qian <[email protected]> |
License: | GPL (>= 2) |
Version: | 0.1-3 |
Built: | 2024-12-22 06:26:05 UTC |
Source: | CRAN |
This provides the solution path of some variants of the adaptive lasso (e.g. SEA-lasso, NSEA-lasso, OLS-adaptive lasso), and the corresponding optimal model selected by BIC criterion.
sealasso(x, y, method = c("nsealasso", "sealasso", "olsalasso", "lasso"))
sealasso(x, y, method = c("nsealasso", "sealasso", "olsalasso", "lasso"))
x |
The model matrix. |
y |
The response. |
method |
One of "nsealasso", "sealasso", "olsalasso" and "lasso", which represent NSEA-lasso, SEA-lasso, OLS-adaptive lasso and the lasso, respectively. The default is "nsealasso". |
SEA-lasso and NSEA-lasso (Qian and Yang, 2010) are two versions of the adaptive lasso. They may be used for variable selection, especially in cases where condition index of the scaled model matrix is large (e.g. > 10) and collinearity is a concern. This function provides condition index, solution path and the suggested optimal model based on BIC. The estimated coefficients are also given for transition points of the path.
method |
The method used. |
condition.index |
Condition index of the scaled model matrix. |
path |
Solution path and corresponding BIC values at transition points. |
beta |
The estimated coefficients at transition points of solution path. |
optim.beta |
The estimated coefficients of the optimal model based on BIC criterion. |
Qian, W. and Yang, Y. (2010) "Model Selection via Standard Error Adjusted Adaptive Lasso." Technical Report, University of Minnesota.
# use the diabetes dataset from "lars" package data(diabetes) x <- diabetes$x y <- diabetes$y sealasso(x, y) # with quadratic terms x2 <- cbind(diabetes$x1,diabetes$x2) object <- sealasso(x2, y, "sealasso") object$condition.index object$optim.beta
# use the diabetes dataset from "lars" package data(diabetes) x <- diabetes$x y <- diabetes$y sealasso(x, y) # with quadratic terms x2 <- cbind(diabetes$x1,diabetes$x2) object <- sealasso(x2, y, "sealasso") object$condition.index object$optim.beta
This provides the method used, condition index and the optimal model selected by BIC.
## S3 method for class 'sealasso' summary(object, ...)
## S3 method for class 'sealasso' summary(object, ...)
object |
a sealasso object |
... |
other arguments to be passed to |
A summary is provided for a sealasso object. To provide a more succinct output, only the method used, condition index and the selected optimal model are printed out.
method |
The method used. One of NSEA-lasso, SEA-lasso, OLS-adaptive lasso and Lasso. |
condition.index |
Condition index of the scaled model matrix. |
optim.beta |
The estimated coefficients of the optimal model based on BIC criterion. |
Qian, W. and Yang, Y. (2010) "Model Selection via Standard Error Adjusted Adaptive Lasso." Technical Report, University of Minnesota.
sealasso
# use the diabetes dataset from "lars" package data(diabetes) y <- diabetes$y # with quadratic terms x2 <- cbind(diabetes$x1,diabetes$x2) object <- sealasso(x2, y, "sealasso") summary(object)
# use the diabetes dataset from "lars" package data(diabetes) y <- diabetes$y # with quadratic terms x2 <- cbind(diabetes$x1,diabetes$x2) object <- sealasso(x2, y, "sealasso") summary(object)