normal: Assumes A Multivariate Normal Prior Distribution for The...

In ehsan66/ICAOD: Optimal Designs for Nonlinear Statistical Models by Imperialist Competitive Algorithm (ICA)

Description

Creates a multivariate normal prior distribution for the unknown parameters as an object of class cprior.

Usage

normal(mu, sigma, lower, upper)

Arguments

mu	A vector representing the mean values.
sigma	A symmetric positive-definite matrix representing the variance-covariance matrix of the distribution.
lower	A vector of lower bounds for the model parameters.
upper	A vector of upper bounds for the model parameters.

Value

An object of class cprior that is a list with the following components:

• fn: prior distribution as an R function with argument param that is the vector of the unknown parameters. See below.

• npar: Number of unknown parameters and is equal to the length of param.

• lower: Argument lower. It has the same length as param.

• upper: Argument lower. It has the same length as param.

The list will be passed to the argument prior of the function bayes. The order of the argument param in fn has the same order as the argument parvars when the model is specified by a formula. Otherwise, it is equal to the argument param in the function fimfunc.

See Also

bayes sensbayes

Examples

normal(mu =  c(0, 1), sigma = matrix(c(1, -0.17, -0.17, .5), nrow = 2),
  lower =  c(-3, .1), upper = c(3, 2))

ehsan66/ICAOD documentation built on Oct. 16, 2020, 8:13 p.m.