auto.ces(data, C=c(1.1, 1), models=c("none","simple","partial","full"), initial=c("backcasting","optimal"), ic=c("AICc","AIC","BIC"), cfType=c("MSE","MAE","HAM","MLSTFE","MSTFE","MSEh"), h=10, holdout=FALSE, intervals=FALSE, level=0.95, intervalsType=c("parametric","semiparametric","nonparametric"), intermittent=c("none","auto","fixed","croston","tsb"), bounds=c("admissible","none"), silent=c("none","all","graph","legend","output"), xreg=NULL, updateX=FALSE, ...)"optimal", meaning that the initial states are optimised, or "backcasting", meaning that the initials are produced using backcasting procedure.
cfType can be: MSE (Mean Squared Error), MAE (Mean Absolute Error), HAM (Half Absolute Moment), MLSTFE - Mean Log Squared Trace Forecast Error, MSTFE - Mean Squared Trace Forecast Error and MSEh - optimisation using only h-steps ahead error. If cfType!="MSE", then likelihood and model selection is done based on equivalent MSE. Model selection in this cases becomes not optimal. There are also available analytical approximations for multistep functions: aMSEh, aMSTFE and aMLSTFE. These can be useful in cases of small samples.
TRUE, the holdout sample of size h will be taken from the data. If FALSE, no holdout is defined.
TRUE, the prediction intervals are constructed.
parametric use state-space structure of ETS. For multiplicative models they are approximated using the same function as for additive. As a result they are a bit wider than should be but are still efficient. In case of mixed models this is done using simulations, which may take longer time than for the pure additive and pure multiplicative models.
semiparametric are based on covariance matrix of 1 to h steps ahead errors and assumption of normal distribution.
nonparametric intervals use values from a quantile regression on error matrix (see Taylor and Bunn, 1999). The model used in this process is e[j] = a j^b, where j=1,..,h.
none, meaning that the data should be considered as non-intermittent; 2. fixed, taking into account constant Bernoulli distribution of demand occurancies; 3. croston, based on Croston, 1972 method with SBA correction; 4. tsb, based on Teunter et al., 2011 method. 5. auto - automatic selection of intermittency type based on data. The first letter can be used instead of the full words.
silent="none", then nothing is silent, everything is printed out and drawn. silent="all" means that nothing is produced or drawn (except for warnings). In case of silent="graph", no graph is produced. If silent="legend", then legend of the graph is skipped. And finally silent="output" means that nothing is printed out in the console, but the graph is produced. silent also accepts TRUE and FALSE. In this case silent=TRUE is equivalent to silent="all", while silent=FALSE is equivalent to silent="none". The parameter also accepts first letter of words ("n", "a", "g", "l", "o").
xreg should have number of observations equal either to in-sample or to the whole series. If the number of observations in xreg is equal to in-sample, then values for the holdout sample are produced using Naive.
TRUE, transition matrix for exogenous variables is estimated, introducing non-linear interractions between parameters. Prerequisite - non-NULL xreg.
FI=TRUE will make the function also produce Fisher Information matrix, which then can be used to calculated variances of parameters of the model.
ces, ets, forecast, ts
y <- ts(rnorm(100,10,3),frequency=12)
# CES with and without holdout
auto.ces(y,h=20,holdout=TRUE)
auto.ces(y,h=20,holdout=FALSE)
library("Mcomp")
## Not run: y <- ts(c(M3$N0740$x,M3$N0740$xx),start=start(M3$N0740$x),frequency=frequency(M3$N0740$x))
# # Selection between "none" and "full" seasonalities
# auto.ces(y,h=8,holdout=TRUE,models=c("n","f"),intervals=TRUE,level=0.8,ic="AIC")## End(Not run)
y <- ts(c(M3$N1683$x,M3$N1683$xx),start=start(M3$N1683$x),frequency=frequency(M3$N1683$x))
test <- auto.ces(y,h=18,holdout=TRUE,intervals=TRUE)
summary(test)
forecast(test)
plot(forecast(test))
Run the code above in your browser using DataLab