A tutorial note on the properties of ARIMA optimal forecasts
Assuming an ARIMA(p,I,q) model represents the data, I show how optimal forecasts can be computed and derive general expressions for its main properties of interest. Namely, I present stepwise derivations of expressions for the variances of forecast errors, and the covariances between them at arbitrary forecasting horizons. Matricial forms for these expressions are also presented to facilitate computational implementation.
|Date of creation:||10 Jan 2012|
|Date of revision:||27 Jul 2012|
|Contact details of provider:|| Postal: Ludwigstraße 33, D-80539 Munich, Germany|
Web page: https://mpra.ub.uni-muenchen.de
More information through EDIRC
When requesting a correction, please mention this item's handle: RePEc:pra:mprapa:40303. See general information about how to correct material in RePEc.
For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: (Joachim Winter)
If references are entirely missing, you can add them using this form.