logistic.enp: Fit a logistic curve to the raw effective number of...
In hisemi: Hierarchical Semiparametric Regression of Test Statistics

Description Usage Arguments Details Value Author(s) See Also

Fit a logistic curve to the raw effective number of parameters over log smoothing parameter

1	logistic.enp(log.spar, enps, maximum, minimum = 2, eps = 1e-08)

`log.spar`	A numeric vector of log smoothing parameters
`enps`	A numeric vector of raw effective number of parameters
`maximum`	A numeric scalar of upper bound of the effective number of parameters
`minimum`	A numeric scalar of lower bound of the effective enumber of parameters
`eps`	A small numeric scalar of error of tolerence

The purpose is of this function is to reduce some estimation variability of effective number of parameters. The function heuristically fit a smooth logistic curve to the raw effective number of parameters.

A numeric vector of fitted effective number of parameters, with attributes

`log.spar`	log of smoothing parameter
`rate`	rate parameter of the logistic curve
`mdis`	the center parameter of the logistic curve
`pow`	the exponent parameter of the logistic curve
`fit`	a nonlinear least squared fit object
`goodenp.idx`	Indices of enps whose values are reasonably estimated
`mode`	The model of enps from a isotonic fit