lambdaChiMCAdjustmentRcpp: Monte Carlo based Group SLOPE tuning parameter correction
In agisga/grpSLOPEMC: Monte Carlo methods for Group SLOPE

Description Usage Arguments Details References

lambdaChiMCAdjustment approximates the variance of (G.10) in Brzyski et. al. (2016) via Monte Carlo, in order to adjust the lambda sequence for correlations in the data.

1 2	lambdaChiMCAdjustmentRcpp(y, X, group_id, lambda, w, number_of_drawings = 5000L)

`y`	The response vector
`X`	The model matrix
`group_id`	A list obtained from `grpSLOPE::getGroupID`
`lambda`	A vector containing the first s entries of lambda
`w`	A vector of weights per group
`number_of_drawings`	The number of iterations in the Monte Carlo procedure

The adjustment is computed for the (s+1)st coefficient of lambda, assuming that the first s coefficients are known. It is required that rank(X) is greater than the sum of the number of elements in any s groups.

D. Brzyski, A. Gossmann, W. Su, M. Bogdan (2016), Group SLOPE - adaptive selection of groups of predictors, https://arxiv.org/abs/1610.04960

agisga/grpSLOPEMC documentation built on May 10, 2019, 7:32 a.m.