simulate.Arima function from the forecast package [closed]

Question

I am using the package forecast for seasonal times series simulations and I have two questions :

1) I don't exactly get the meaning and the utility of the "future" option. It is set to TRUE by default, and I think it should be so if I want to predict future values of the series, but I don't understand what is the use of a simulation with future=FALSE.

2) The simulate.Arima function is basically an improvement for the traditionnal arima.sim one. However with arima.sim, it is possible to provide some user-defined innovations processes to the function using the innov argument while it is not possible to do so with simulate.Arima. Did I miss something ? If not, and if Mr Hyndman reads this post, could it be possible to add such an option in a future release ? For the moment, I think I'll get the source code and try to modify the code by myself.

Thanks and have a nice day.

Answer 1

1. According to the help file, future means "Produce sample paths that are future to and conditional on the data in object.". So if future=TRUE, the simulated observations are conditional on the historical observations. In other words, they are possible future sample paths of the time series. But if future=FALSE, the historical data are ignored, and the simulations are possible realizations of the time series model that are not connected to the original data.

2. I will add the suggestion of allowing user-specified innovations to the list of feature requests at https://github.com/robjhyndman/forecast/issues?state=open. In the meantime, it is a very easy modification. Just find the call to rnorm and replace it.