hessianReg: Hessian Matrix of Regularized Estimators

Description Usage Arguments Details Value Author(s) References See Also Examples

View source: R/zipath.R

Description

Constructing Hessian matrix for regularized regression parameters.

Usage

1

Arguments

x

a fitted model object.

which

which penalty parameter(s)?

...

arguments passed to the meatReg function.

Details

hessianReg is a function to compute the Hessian matrix estimate of non-zero regularized estimators. Implemented only for zipath object with family="negbin" in the current version.

Value

A matrix containing the Hessian matrix estimate for the non-zero parameters.

Author(s)

Zhu Wang <[email protected]>

References

Zhu Wang, Shuangge Ma and Ching-Yun Wang (2015) Variable selection for zero-inflated and overdispersed data with application to health care demand in Germany, Biometrical Journal. 57(5):867-84.

See Also

breadReg, meatReg

Examples

1
2
3
data("bioChemists", package = "pscl")
fm_zinb <- zipath(art ~ . | ., data = bioChemists, family = "negbin", nlambda=10)
hessianReg(fm_zinb, which=which.min(fm_zinb$bic))

Example output

              [,1]          [,2]          [,3]          [,4]         [,5]
[1,] -8.594523e+02 -3.706159e+02 -4.075601e+02 -8.757817e+03  0.073177587
[2,] -3.706159e+02 -3.706159e+02 -8.642319e+01 -3.271976e+03  0.033817541
[3,] -4.075601e+02 -8.642319e+01 -6.641264e+02 -4.435521e+03  0.038678552
[4,] -8.757817e+03 -3.271976e+03 -4.435521e+03 -1.798840e+05  0.538212523
[5,]  7.317759e-02  3.381754e-02  3.867855e-02  5.382125e-01 -0.003377995
[6,]  5.382125e-01  2.173251e-01  2.898499e-01  8.335621e+00 -0.188207596
[7,]  1.616354e+00 -3.399707e+00 -5.487150e+00  4.862150e+01  0.009792022
            [,6]          [,7]
[1,]  0.53821252   1.616354009
[2,]  0.21732514  -3.399706842
[3,]  0.28984991  -5.487149661
[4,]  8.33562112  48.621496304
[5,] -0.18820760   0.009792022
[6,] -5.73187787   0.086554221
[7,]  0.08655422 -14.645095376

mpath documentation built on Aug. 19, 2017, 1:03 a.m.