I have a double seasonal data. I wrote the following code to find the best fit model and find fitted values:

orders <- read.csv("DataForR.csv", header = TRUE), NumOrders <- orders$Orders, > period <- orders$Period, OrderTimeSeries <- msts(NumOrders,seasonal.periods=c(24,7*24)), fit2 <- dshw(OrderTimeSeries,24,168,336,alpha=NULL,beta=NULL,gamma=NULL,omega=NULL,phi=NULL,lambda=NULL,armethod=TRUE,model=NULL),

my original time series plot is as follows: enter image description here

but the fitted values are as follows:

enter image description here

I wonder what does a fitted value mean? does it mean using the fitted model, the forecast values for the training dataset are fitted data? how can I have both series in one plot so I can better compare their trends? Thanks


The fitted values correspond to an optimal 1-step ahead forecast.

See: http://cran.ms.unimelb.edu.au/web/packages/forecast/forecast.pdf


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.