Hi Akshay,
The forecast package will do the BoxCox transform and automatically
backtransform the forecasts.
The package also handles xts objects.
For example, modifying the example from the help page of
forecast::forecast for Arima
> dt <- as.Date("2023-01-01") + 1:length(WWWusage)
> a <- xts(WWWusage, order.by=dt)
> fit1 <- Arima(a, c(3,1,0))
> fit2 <- Arima(a, lambda=0.5, c(3,1,0)) ## applies the Box-Cox transform
with lambda=0.5
> par(mfrow=c(1,2))
> plot(forecast(fit1))
> plot(forecast(fit2))
HTH,
Eric
p.s. RJH is the author/maintainer of the forecast package
On Sun, Aug 13, 2023 at 1:01?AM akshay kulkarni <akshay_e4 at hotmail.com>
wrote:>
> dear members,
> I have a heteroscedastic time series which I want
to transform to make it homoscedastic by a box cox transformation. I am using
Otexts by RJ hyndman and George Athanopolous as my textbook. They discuss
transformation and also say the fpp3 and the fable package automatically back
transforms the point forecast. they also discuss the process which I find to be
very cumbersome. Is there any R package which automatically back transforms the
point forecast when I use xts objects ( RJH and GA use tsibble objects) with
arfima/arima in the forecast package?
>
> THanking you,
> Yours sincerely,
> AKSHAY M KULKARNI
>
> [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.