cdfPen.fit {penalizedcdf}

Fitter function for CDF penalty

Description

These are the fundamental computing algorithms that cdfPen invokes to estimate penalized linear models by varying lambda.

Usage

cdfPen.fit(b,
           b.tld,
           g,
           b.rho,
           H.rho,
           lmb.rho,
           nu,
           algorithm,
           nstep = 1E+5,
           eps = 1E-5,
           eps.lla = 1E-6,
           nstep.lla = 1E+5)

Arguments

b

Starting values of beta-vector.

b.tld

Starting values of sparse beta-vector.

g

Starting values of pseudo-variable.

b.rho

Ridge solution.

H.rho

Second part of ridge solution.

lmb.rho

Lambda-rho ratio.

nu

Shape parameter of the penalty. It affects the degree of the non-convexity of the penalty.

algorithm

Approximation to be used to obtain the sparse solution.

nstep

Maximum number of iterations of the global algorithm.

eps

Convergence threshold of the global algorithm.

eps.lla

Convergence threshold of the LLA-algorithm (if used).

nstep.lla

Maximum number of iterations of the LLA-algorithm (if used).

Value

b

Estimated beta-vector.

b.tld

Estimated sparse beta-vector.

g

Final values of pseudo-variable.

i

Number of iterations.

conv

Convergence check status (0 if converged).

Author(s)

Daniele Cuntrera, Luigi Augugliaro, Vito Muggeo

References

Aggiungere Arxiv