vec2par: Convert a Vector of Parameters to a Parameter Object of a Distribution

Description

This function converts a vector of parameters to a parameter object of a distribution. The type of distribution is specified in the argument list: aep4, cau, exp, gam, gep, gev, glo, gno, gpa, gum, kap, kur, lap, lmrq, ln3, nor, pe3, ray, revgum, rice, st3, texp, wak, and wei. These abbreviations and only these are used in routing logic within lmomco. There is no provision for fuzzy matching. However, if the distribution type is not identified, then the function issues a warning, but goes ahead and creates the parameter list and of course can not check for the validity of the parameters. If one has a need to determine on-the-fly the number of parameters in a distribution as supported in lmomco, then see the dist.list function.

Usage

vec2par(vec, type, nowarn=FALSE, paracheck=TRUE, ...)

Arguments

vec

A vector of parameter values for the distribution specified by type.

type

Three character distribution type (for example, type='gev').

nowarn

A logical switch on warning suppression. If TRUE then options(warn=-1) is made and restored on return. This switch is to permit calls in which warnings are not desired as the user knows how to handle the returned value---say in an optimization algorithm.

paracheck

A logical controlling whether the parameters and checked for validity. Overriding of this check might be extremely important and needed for use of the distribution quantile function in the context of TL-moments with nonzero trimming.

...

Additional arguments for the are.par.valid call that is made internally.

Value

An R list is returned. This list should contain at least the following items, but some distributions such as the revgum have extra.

Details

If the distribution is a Reverse Gumbel (type=revgum) or Generalized Pareto (type=gpa), which are 2-parameter or 3-parameter distributions, the third or fourth value in the vector is the $\zeta$ of the distribution. $\zeta$ represents the fraction of the sample that is noncensored, or number of observed (noncensored) values divided by the sample size. The $\zeta$ represents censoring on the right, that is there are unknown observations above a threshold or the largest observed sample. Consultation of parrevgum or pargpaRC should elucidate the censoring discussion.

Examples

Run this code

para <- vec2par(c(12,123,0.5),'gev')
Q <- quagev(0.5,para)

my.custom <- vec2par(c(2,2), type='myowndist') # Think about making your own