search

robjhyndman / forecast

Forecasting Functions for Time Series and Linear Models
http://pkg.robjhyndman.com/forecast
1.13k stars 342 forks source link

Add arimax() #7

Closed robjhyndman closed 6 years ago

robjhyndman commented 12 years ago

I would like transfer function methodology to be added. We need to decide whether to use the Peter Young approach, as in the CAPTAIN toolbox in Matlab, or a state space approach as in Brockwell & Davis, or something else.

BrianMiner commented 9 years ago

I would LOVE to see this. Given some of the weaknesses with the TSA package, there is essentially nothing in R that allows building intervention / interrupted time series models. Given how robust 'forecast' has become, it would be the perfect place!

ekote commented 8 years ago

Hi @robjhyndman I would like to implement ARIMAX in open source Spark-timeseries (scala) library because it hasn't existed yet. You wrote:

The arimax() function from the TSA package fits the transfer function model (but not the ARIMAX model).

Could you explain how the transfer function model work and how should ARIMAX be pragmatically implemented? Or any research on which I can base on my code. Thank you! Estera

robjhyndman commented 8 years ago

@ekote . This is not the place for a general discussion. Ask on crossvalidated.com.