momentuHMM (version 1.4.3)

AICweights: Calculate Akaike information criterion model weights

Description

Calculate Akaike information criterion model weights

Usage

AICweights(..., k = 2, n = NULL)

Arguments

...

momentuHMM, HMMfits, or miHMM objects, to compare AIC weights of the different models.

k

Penalty per parameter. Default: 2 ; for classical AIC.

n

Optional sample size. If specified, the small sample correction AIC is used (i.e., AICc = AIC + kp(p+1)/(n-p-1) where p is the number of parameters).

Value

The AIC weights of the models. If multiple imputation objects are provided, then the mean model weights (and standard deviations) are provided.

Details

• Model objects must all be either of class momentuHMM or multiple imputation model objects (of class HMMfits and/or miHMM).

• AIC is only valid for comparing models fitted to the same data. The data for each model fit must therefore be identical. For multiple imputation model objects, respective model fits must have identical data.

Examples

# NOT RUN {
# }
# NOT RUN {
# HMM specifications
nbStates <- 2
stepDist <- "gamma"
angleDist <- "vm"
mu0 <- c(20,70)
sigma0 <- c(10,30)
kappa0 <- c(1,1)
stepPar0 <- c(mu0,sigma0)
anglePar0 <- c(-pi/2,pi/2,kappa0)
formula <- ~cov1+cov2
                
# example$m is a momentuHMM object (as returned by fitHMM), automatically loaded with the package
mod1 <- fitHMM(example$m$data,nbStates=nbStates,dist=list(step=stepDist,angle=angleDist),
                Par0=list(step=stepPar0,angle=anglePar0),
                formula=~1,estAngleMean=list(angle=TRUE))

Par0 <- getPar0(mod1,formula=formula)                 
mod2 <- fitHMM(example$m$data,nbStates=nbStates,dist=list(step=stepDist,angle=angleDist),
                Par0=Par0$Par,beta0=Par0$beta,
                formula=formula,estAngleMean=list(angle=TRUE))
                
AICweights(mod1,mod2)
# }
# NOT RUN {
# }