forecastHybrid
forecastHybrid copied to clipboard
Adding ARFIMA models.
I would be nice to add the arfima
model from forecast package.
Such "model" are quite uncommon but are working quite well for times series that have long rum memory such as consumer/Producer price indexes. Especially in some countries where there is mid-month divulgations.
I think it would be quite easy, just changing a few lines of code.
models <- c("a" = auto.arima, "e" = ets, "f" = thetam, "n" = nnetar,
"s" = stlm, "t" = tbats, "z" = snaive) "s" = stlm, "t" = tbats, "z" = snaive, "x"=arfima)
validModels <- c("a", "e", "f", "n", "s", "t", "z", "x")
It is working on my computer. The only problem is an error message when I run the cvts
function with a hybridModel
that include a "x" component.
Error in { : task 1 failed - "'data' must be of a vector type, was 'NULL'"
Any ideas ?