waic: Function for the Watanabe-Akaike information criteria (WAIC)

In artemis-toumazi/dfped: Extrapolation and Bridging of Adult Information in Early Phase Dose-Finding Paediatrics Studies

Description

Model selection can be performed for each working model (WM) using the Watanabe-Akaike information criteria (WAIC) developed by Watanabe.

Usage

waic(stanfit, s)            

Arguments

stanfit	Estimates obtained with the STAN fit. You can use the fitDataj function which is giving the next fit of the model from STAN.
s	Integer specifying the number of models used to compute the WAIC selection.

Author(s)

Artemis Toumazi [e-mail], Caroline Petit [e-mail], Sarah Zohar [e-mail]

References

Petit, C., et al, (2016) Unified approach for extrapolation and bridging of adult information in early phase dose-finding paediatric studies, Statistical Methods in Medical Research, <doi:10.1177/0962280216671348>.

Watanabe S. Asymptotic Equivalence of Bayes cross vallidation and widely applicable information criterion in singular learning theory, volume 11. 2010.

Examples

## Not run: 
	for(s in 1:nbDesign){
		fitj <- fitDataj(stan_model, nbPatientsj, nbDoses, tox, eff, given_dose,
						 skeleton_tox, skeleton_eff, mu, sigma, s)
		waicj <- waic(stanfit=fitj, s)
	}

## End(Not run)

artemis-toumazi/dfped documentation built on May 10, 2019, 1:49 p.m.