expWithConst: Exponentiate a vector of values plus a constant

In hypergsplines: Bayesian model selection with penalised splines and hyper-g prior

Description

This functions tries to choose the constant such that the smallest result is larger than zero and the largest result is smaller than infinity, in the numerical representation. If this is not possible, results numerically identical to zero are tolerated. Only values which are finite are taken into account for this considerations, because infinite values will be either zero or infinity anyway, regardless of the constant.

Usage

  expWithConst(values)

Arguments

values

argument values, of which at least one must be finite

Value

the exp(values + constant) results, see details

Author(s)

Daniel Sabanes Bove

hypergsplines documentation built on May 2, 2019, 6:14 p.m.