empirical.GoF: Empirical Goodnesses of Fit

In hkauhanen/pbcm: Parametric Bootstrap Cross-Fitting Method

Description

Fit models 1 and 2 to data and return the empirical goodnesses of fit as well as the difference in goodness of fit.

Usage

empirical.GoF(
  data,
  fun1,
  fun2,
  args1 = NULL,
  args2 = NULL,
  verbose = TRUE,
  GoFname = "GoF"
)

Arguments

data	Data frame
fun1	Modelling function 1
fun2	Modelling function 2
args1	List of arguments passed to fun1
args2	List of arguments passed to fun2
verbose	If TRUE, warnings are printed to the console
GoFname	Name of the element returned by fun1 and fun2 holding the goodness of fit

Details

Functions fun1 and fun2 must accept data as an argument in addition to any arguments specified in args1 and args2. They must return a list with an element carrying the calculated goodness of fit; by default the name of this element is taken to be the string "GoF" but this behaviour can be changed through the GoFname argument.

Value

A 1-row data frame of three columns:

GoF1

Goodness of fit for model 1

GoF2

Goodness of fit for model 2

DeltaGoF

Equal to GoF1 - GoF2

Author(s)

Henri Kauhanen

Examples

x <- seq(from=0, to=1, length.out=100)
mockdata <- data.frame(x=x, y=x + rnorm(100, 0, 0.5))

myfitfun <- function(data, p) {
  res <- nls(y~a*x^p, data, start=list(a=1.1))
  list(a=coef(res), GoF=deviance(res))
}

empirical.GoF(mockdata, fun1=myfitfun, fun2=myfitfun,
              args1=list(p=1), args2=list(p=2))

hkauhanen/pbcm documentation built on July 6, 2020, 12:25 a.m.