AIC.envcpt: Computes the AIC measure for output from the envcpt function.

Description

Uses the likelihood and number of parameters from the output of the envcpt function and calculates the AIC measure for each model.

Usage

# S3 method for envcpt
AIC(object,...,k=2)

Arguments

object

A list produced as output from the envcpt function. In essence any list where the first element contains a matrix whose first row is twice the negative log-likelihood and second row is the number of parameters. Columns are different models to compare.

...

Not used

numeric, the penalty per parameter to be used: the default k=2 is the classical AIC.

Value

Vector of AIC values the same length as the number of columns in the first entry to the input list (length 8 if output from envcpt is used). The column names from the envcpt output are preserved to give clear indication on models.

Details

Calculates the AIC defined as -2*<log-likelihood> + 2*<number of parameters>. When comparing models the smaller the AIC the better the fit.

Examples

Run this code

# NOT RUN {
set.seed(1)
x=c(rnorm(100,0,1),rnorm(100,5,1))
out=envcpt(x) # run the 8 models with default values
out[[1]] # first row is twice the negative log-likelihood for each model
         # second row is the number of parameters
AIC(out) # returns AIC for each model.
which.min(AIC(out)) # gives meancpt (model 2) as the best model fit.
out[[3]] # gives the model fit for the meancpt model.
plot(out,type='fit') # plots the fits
plot(out,type="aic") # plots the aic values

set.seed(10)
x=c(0.01*(1:100),1.5-0.02*((101:250)-101))+rnorm(250,0,0.2)
out=envcpt(x,minseglen=10) # run the 8 models with a minimum of 10 observations between changes
AIC(out) # returns the AIC for each model
which.min(AIC(out)) # gives trendcpt (model 6) as the best model fit.
out[[7]] # gives the model fit for the trendcpt model.
plot(out,type='fit') # plots the fits
plot(out,type="aic") # plots the aic values

set.seed(100)
x=arima.sim(model=list(ar=0.8),n=100)+5
out=envcpt(x) # run the 8 models with 
AIC(out) # returns the AIC for each model
which.min(AIC(out)) # gives trendar (model 7) as the best model fit.
out[[7]] # gives the model fit for the trendar model. Notice that the trend is tiny but does 
# produce a significantly better fit than the meanar model.
plot(out,type='fit') # plots the fits
plot(out,type="aic") # plots the aic values
# }