Page 1 of 1

arima for daily data

Posted: Mon Feb 17, 2014 3:20 pm
by marijask
Hi, Im new in eviews and need a help for forecasting daily data with arima (univariate model for CiC).
What should I do first? create a dummy variables (for day of the week, months, moving holidays seasonalities), and than specify the arima and make a model?
Is there any code in eviews for such arima model for daily data?
Thnx in advance

Re: arima for daily data

Posted: Mon Apr 14, 2014 5:41 am
by BobJ
You would do ARMA modelling similarly to the way you'd do it with data at other frequencies. Box-Jenkins identification (using autocorrelation and partial autocorrelations) to select the optimal ARMA model and then you estimate the chosen model.There's no specific code for that in eviews. You have to it manually.