# maxLGlasso: An algorithm for determining the smallest values for Lasso... In RegularizedSCA: Regularized Simultaneous Component Based Data Integration

## Description

`maxLGlasso` identify the minimum value for Lasso and Group Lasso tuning parameters that lead to an estimated P matrix with all of its elements equal 0. This minimum value is thus the maximum value (the boundary) that users should consider for Lasso and Group Lasso. Note that the algorithm is based on the "component" method; see sparseSCA.R

## Usage

 `1` ```maxLGlasso(DATA, Jk, R) ```

## Arguments

 `DATA` The concatenated data block, with rows representing subjects. `Jk` A vector. Each element of this vector is the number of columns of a data block. `R` The number of components.

## Value

 `Glasso` The maximum value for Group Lasso tuning parameter. `Lasso` The maximum value for Lasso tuning parameter.

## Note

The description of how to obtain the maximum value for Lasso tuning parameter can be found in page 17 of Hastie, Tibshirani, and Wainwright (2015). We are not aware of any literature that mentions how to obtain the maximum value for Group Lasso, but this value can easily be derived from the algorithm.

## References

Hastie, T., Tibshirani, R., & Wainwright, M. (2015). Statistical learning with sparsity. CRC press.

## Examples

 ``` 1 2 3 4 5 6 7 8 9 10``` ```## Not run: DATA1 <- matrix(rnorm(50), nrow=5) DATA2 <- matrix(rnorm(100), nrow=5) DATA <- cbind(DATA1, DATA2) Jk <- c(10, 20) results <- maxLGlasso(DATA, Jk, R=5) maxGLasso <- results\$Glasso maxLasso <- results\$Lasso ## End(Not run) ```

RegularizedSCA documentation built on May 2, 2019, 8:24 a.m.