pomp (version 0.30-1)

slice.design: Design matrices for likelihood slices and profiles

Description

slice.design generates a data-frame representing points taken along one or more slices through a point in a multidimensional space. profile.design generates a data-frame where each row can be used as the starting point for a profile likelihood calculation.

Usage

slice.design(vars, n)
profile.design(..., vars, n)

Arguments

...

Specifies the parameters over which to profile.

vars

For the slice.design case: vars is a named list of numeric vectors, each of which has length either 1 or 3. Variables along which slices are to be taken should have length 3, corresponding to the minimum of the range, cen

n

In the case of slice.design, the number of points per slice. In the case of profile.design, the number of starts per profile point.

Value

• slice.design returns a data frame with n points per slice. The column slice is a factor that tells which slice each point belongs to.

  profile.design returns a data frame with n points per profile point. The parameters in vars are sampled using sobol.

See Also

sobol

Examples

## A single 11-point slice through the point c(A=3,B=8,C=0) along the B direction.
x <- slice.design(list(A=3,B=c(0,8,10),C=0),n=11)
dim(x)
plot(x)
## Two slices through the same point along the A and C directions.
x <- slice.design(list(A=c(0,3,5),B=8,C=c(0,0,5)),n=11)
dim(x)
plot(x)

## A one-parameter profile design:
x <- profile.design(p=1:10,vars=list(a=c(0,1),b=c(0,5)),n=20)
dim(x)
plot(x)
## A two-parameter profile design:
x <- profile.design(p=1:10,q=3:5,vars=list(a=c(0,1),b=c(0,5)),n=20)
dim(x)
plot(x)