F.start.limits: F.start.limits - Set starting values and limits for parameters of Rdistance functions.

Description

Return reasonable starting values and limits (boundaries) for the parameters of distance functions. Starting values and limits are specified for all likelihoods and expansion terms. This function is called by other routines in Rdistance, and is not intended to be called by the user.

Usage

F.start.limits(like, expan, w.lo, w.hi, dist)

Arguments

String specifying the likelihood for the distance function. Possible values are "hazrate" for hazard rate likelihood, "halfnorm" for the half normal likelihood, "uniform" for the uniform likelihood, "negexp" for the negative exponential

expan

Number of expansion terms to include. Valid values are 0, 1, ..., 3.

w.lo

Lower or left-truncation limit of the distances. Normally, 0.

w.hi

Upper or right-truncation limit of the distances. This is the maximum off-transect distance that could be observed.

dist

The vector of observed off-transect distances being analyzed. This vector is only required for like = "Gamma" and "halfnorm".

Value

A list containing the following components
startVector of reasonable starting values for parameters of the likelihood and expansion terms.
lowlimitVector of lower limits for the likelihood parameters and expansion terms.
uplimitVector of upper limits for the likelihood parameters and expansion terms.
namesVector of names for the likelihood parameters and expansion terms.

Details

The number of parameters to be fitted is expan + 1 + 1*(like %in% c("hazrate", "uniform")). This is the length of all vectors returned in the output list.

Examples

Run this code

F.start.limits( "uniform", 0, 0, 1000 )
    F.start.limits( "uniform", 1, 0, 1000 )
    F.start.limits( "uniform", 2, 0, 1000 )
    F.start.limits( "uniform", 3, 0, 1000 )

    F.start.limits( "halfnorm", 0, 0, 1000, 500*runif(100) )
    F.start.limits( "halfnorm", 1, 0, 1000, 500*runif(100) )
    F.start.limits( "halfnorm", 2, 0, 1000, 500*runif(100) )
    F.start.limits( "halfnorm", 3, 0, 1000, 500*runif(100) )

    F.start.limits( "hazrate", 0, 0, 1000 )
    F.start.limits( "hazrate", 1, 0, 1000 )
    F.start.limits( "hazrate", 2, 0, 1000 )
    F.start.limits( "hazrate", 3, 0, 1000 )

    F.start.limits( "negexp", 0, 0, 1000 )
    F.start.limits( "negexp", 1, 0, 1000 )
    F.start.limits( "negexp", 2, 0, 1000 )
    F.start.limits( "negexp", 3, 0, 1000 )

    F.start.limits( "Gamma", 0, 0, 1000, 1000*runif(100) )

Run the code above in your browser using DataLab

Last chance! 50% off unlimited learning