fdesigns (version 1.1)

fdesigns-package: tools:::Rd_package_title("fdesigns")

Description

tools:::Rd_package_description("fdesigns")

Author

tools:::Rd_package_author("fdesigns")

Maintainer: tools:::Rd_package_maintainer("fdesigns")

Details

The DESCRIPTION file: tools:::Rd_package_DESCRIPTION("fdesigns") The most important functions are pflm and pfglm which can be used to identify optimal designs for functional linear and generalised linear models, respectively, using the coordinate exchange algorithm.

References

Michaelides, D. (2023). Design of experiments for models involving profile factors (Doctoral dissertation, University of Southampton).

Examples

## Example 1:
## This example involves finding an A-optimal design for a functional linear model of 4 runs
## depending on one profile factor. The settings of the profile factor are represented by a 
## B-spline basis of degree zero and a single knot at (0.5). The single functional parameter 
## is represented with a linear power series basis. Five random starts are chosen.

example1 <- pflm(formula = ~ x1, nsd = 5, mc.cores = 1, npf = 1,
  tbounds = c(0, 1), nruns = 4, startd = NULL,  dx = c(0), 
  knotsx = list(c(0.5)), pars = c("power"), db = c(1), 
  knotsb = list(c()), criterion = "A", lambda = 0, 
  dlbound = -1, dubound = 1, tol = 0.0001, progress = FALSE)

print(example1) ##  prints the output of example1.
##
## The number of profile factors is: 1
##
## The number of runs is: 4
##
## The objective criterion is: A-optimality
##
## The objective value is: 8.75
##
## The number of iterations is: 6
##
## The computing elapsed time is: 00:00:00


## Example 2:
## This example involves finding an A-optimal design for a functional logistic
## model of 12 runs depending on one profile factor. The settings of the profile 
## factor are represented by a B-spline basis of degree zero and a three interior knots 
## at (0.25, 0.50, 0.75). The single functional parameter is represented with a linear 
## power series basis. The method of approximation is Monte Carlo with the prior 
## specified by the function prmc. Three random starts are chosen.

set.seed(100) ## Set seed to achieve reproducibility.

prmc <- function(B,Q) {
  matrix(rnorm(B*Q, mean=0, sd=sqrt(2)), nrow=B, ncol=Q)
}
## A function which specifies the prior. This function returns a 
## B by Q matrix of randomly generated values from the prior 
## distribution for the model parameters.

# \donttest{
example2 <- pfglm(formula = ~ 1 + x1, nsd = 3, mc.cores = 1, npf = 1, 
             tbounds = c(0,1), nruns = 12, startd = NULL, 
             dx = c(0), knotsx = list(c(0.25,0.50,0.75)), 
             pars = c("power"), db = c(1), knotsb = list(c()), 
             lambda = 0, criterion = "A", family = binomial,
             method=c("MC"), level = 6, B = 10000, prior = prmc,
             dlbound = -1, dubound = 1, tol = 0.0001, progress = TRUE)

print(example2) ##  prints the output of example1.
##
## The number of profile factors is: 1
##
## The number of runs is: 12
##
## The objective criterion is: A-optimality
##
## The objective value is: 20.23283
##
## The number of iterations is: 5
##
## The method of approximation is: MC
##
## The family distribution and the link function are: binomial and logit
##
## The computing elapsed time is: 00:00:12
# }