Learn R Programming

MortalitySmooth (version 2.3.4)

Mort1Dsmooth_optimize: Optimize a 1D Penalized-Poisson IWLS over smoothing parameters

Description

This is an internal function of package MortalitySmooth which optimizes the smoothing parameter for penalized B-splines within the function Mort1Dsmooth.

Usage

Mort1Dsmooth_optimize(x, y, offset, wei, psi2, B, DtD, a.init, MON, TOL1, TOL2, RANGE, MAX.IT, MET)

Arguments

x
vector for the abscissa of data.
y
vector of counts response.
offset
vector with an a priori known component (optional).
wei
an optional vector of weights to be used in the fitting process.
psi2
an overdispersion parameter used the quasi-likelihood approach.
B
B-splines basis.
DtD
inner product of the difference matrix
a.init
initial coefficients
MON
Logical switch indicating if monitoring is required.
TOL1
The tolerance level in the IWLS algorithm.
TOL2
difference between two adjacent smoothing parameters in the (pseudo) grid search, log-scale.
RANGE
range in which smoothing parameter should be searched.
MAX.IT
the maximum number of iterations
MET
the method for controlling the amount of smoothing

Details

The function aims to find the optimal smoothing parameter within the given RANGE in Mort1Dsmooth with method equal to 1 or 2 (BIC and AIC). It employs the function cleversearch from package svcm in two separate steps. First it searches using a rough grid (4 times TOL2) and the median of RANGE as starting lambda. Afterwards it searches in the restricted range around the sub-optimal smoothing parameter, using a finer grid defined by TOL2.

This procedure allows to find a precise smoothing parameter in an efficient way: we do not explore the full range of possible lambda values, moving at most one grid step up or down. Furthermore the two steps routine reduces the risk of finding sub-optimal smoothing parameter.

References

Camarda, C. G. (2012). MortalitySmooth: An R Package for Smoothing Poisson Counts with P-Splines. Journal of Statistical Software. 50, 1-24. http://www.jstatsoft.org/v50/i01/.

See Also

Mort1Dsmooth_update, Mort1Dsmooth_estimate, Mort1Dsmooth.