evaluateFixedDesigns: Evaluate fixed designs

In apear9/SSNdesign: An R Package for Optimal and Adaptive Sampling Designs on Stream Networks

Description

Evaluates the utility of a list of user-specified designs.

Usage

evaluateFixedDesigns(ssn, glmssn, list.designs, list.of = "pid",
  utility.function, prior.parameters, n.draws = 1000,
  extra.arguments = NULL)

Arguments

ssn	an object of class SpatialStreamNetwork
glmssn	an object of class glmssn
list.designs	a list containing vectors of pid or locID values. Use pid values if the sites in a design change over time.
list.of	a string (either "pid" or "locID") indicating whether the designs are expressed as vectors of pid or locID values
utility.function	a function with the signature utility.function. Users may define their own. This package provides several built-in utility functions.
prior.parameters	a function to act as a prior for covariance parameter values
n.draws	a numeric value for the number of Monte Carlo draws to take when evaluating potential designs. This defaults to 1000.
extra.arguments	a list of any extra parameters which can be used to control the behaviour of this function or the utility function

Details

This function takes a list of user-specified designs for a spatial stream network and evaluates some utility function over those designs. This function should be relatively fast compared to the functions which search for an optimal design. However, because this function is not parallellised, it is recommended that users limit themselves to fewer than one hundred designs. Note, this function is called internally in the design diagnostic functions.

Value

A data.frame with the columns: ID (integers from 1 in the same order as list.designs); Size, the number of design points; Expected Utility; Efficiency, the ratio between each expected utility and the largest expected utility; and Efficiency_Unlogged, the same but for expected utilities expressed on the log scale.

apear9/SSNdesign documentation built on Feb. 19, 2020, 4:29 a.m.