CPO: Conditional predictive ordinates

In oswaldosantos/INLAOutputs: Process selected outputs from the 'INLA' package

Description

Sort and display the sum of the log of CPO's of INLA models.

Usage

CPO(..., failure_cutoff = 0, decreasing = TRUE, rnd = 3)

Arguments

...	objects of class inla with computed CPO's.
failure_cutoff	value between 0 and 1. CPO's with a failure score greater than failure_cutoff are considered unreliable and are ignored.
decreasing	logical. If TRUE (default), CPO's are displayed in decreasing order (higher values indicate better prediction).
rnd	integer indicating the number of decimal places to be used.

Details

control.compute = list(cpo = TRUE) must be used within inla function.

Value

matrix with the sum of the log of CPO's and the proportion of failures of each models' CPO.

References

Blangiardo, M., Cameletti, M., Baio, G., & Rue, H. (2013). Spatial and spatio-temporal models with R-INLA. Spatial and spatio-temporal epidemiology, 7, 39-55.

Examples

data(sp)

mod1 <- inla(aan ~ f(id, model = 'bym', graph = sp.adj),
             family = 'poisson', data = spn, E = eaan,
             control.compute = list(cpo = TRUE))

mod2 <- inla(aan ~ shvn + f(id, model = 'bym', graph = sp.adj),
             family = 'poisson', data = spn,  E = eaan,
             control.compute = list(cpo = TRUE))

CPO(mod1, mod2)

oswaldosantos/INLAOutputs documentation built on May 24, 2019, 5 p.m.