momApp: Moment approximation for polynomial PDMPs

Description Usage Arguments Value Note See Also Examples

Description

This function calculates the raw moments of a polynomial PDMP. The calculation is not based on simulated data but uses the generator of the PDMP to create a system of ordinary differential equations. The moments can be calculated from the solution of this system of ode's.

In some cases, the system of ode's is infinetly large and can therefore not be solved. A modification is necessary, meaning that all equations for moments whose order is greater than a given order will be replaced by fixed values. This process is called moment closure. There are different possibilities, which values to set. They depend on the parameter closure. The following options are implemented:

Parameter centralize determines, if raw or central moments should be replaced by fixed values.

Usage

1
2
3
4
5
6
7
8
momApp(obj, maxorder = 4, na.rm = TRUE, closure = c("zero", "zero",
  "normal", "lognormal", "gamma"), centralize = c(TRUE, FALSE, TRUE,
  FALSE, FALSE))

## S4 method for signature 'polyPdmpModel'
momApp(obj, maxorder = 4, na.rm = TRUE,
  closure = c("zero", "zero", "normal", "lognormal", "gamma"),
  centralize = c(TRUE, FALSE, TRUE, FALSE, FALSE))

Arguments

obj

object of class polyPdmpModel.

maxorder

integer defining the highest order of moments that are considered. Higher orders are droped and replaced by fixed values. The replacement method is specified in parameter closure.

closure

character vector. Every entry defines one possibility of moment closure, i.e. changing the system of ODEs into a closed form that is solvable. Possible values are zero, normal, lognormal or gamma.

centralize

boolean vector with the same length as closure. Entry centralize[i] defines if the replacement of moments with fixed values according to closure[i] should be performed on raw or central moments.

Value

The function returns an S3-object of class momApp. It contains 6 different elements:

Note

This method works only for PDMPs with one discrete variable.

See Also

momApp-methods for further analysing the result.

Examples

1
2
3
4
5
data(genePolyBF)
a <- momApp(genePolyBF, maxorder = 4)
plot(a)
print(a)
summary(a)

CharlotteJana/pdmppoly documentation built on Sept. 4, 2019, 4:40 p.m.