LoglikPen

A p-dimensional vector containing the regression ceofficients in the CoxPH model.

beta

time

The status indicator: 1 for event and 0 for censoring.

status

The penalty parameter euqals either 2 in AIC or ln(n0) in BIC (by default), where n0 is the number
of uncensored survival times observed in the data. You can also specify it to a specific value of your own choice.

lambda

The scale parameter in the hyperbolic tangent function of the MIC penalty. By default, \(a = n0\), i.e., the number
of uncensored survival times observed in the data.

Compute the penalized log partial likelihood for a Cox PH model with MIC penalty

Sparse estimation for Cox PH models is done via
Minimum approximated Information Criterion (MIC) by Su, Wijayasinghe,
Fan, and Zhang (2016) <DOI:10.1111/biom.12484>. MIC mimics the best
subset selection using a penalized likelihood approach yet with no need
of a tuning parameter. The problem is further reformulated with a
re-parameterization step so that it reduces to one unconstrained non-convex
yet smooth programming problem, which can be solved efficiently. Furthermore,
the re-parameterization tactic yields an additional advantage in terms of
circumventing post-selection inference.

LoglikPen: Compute the penalized log partial likelihood for a Cox PH model with MIC penalty

Description

Usage

Arguments

Value

References

See Also