Computing the best linear predictor in a Hilbert space. Applications to general ARMAH processes

作者：Bosq D^*

来源：Journal of Multivariate Analysis, 2014, 124: 436-450.

DOI：10.1016/j.jmva.2013.11.013

• Functional filters
• Hilbert spaces
• Linear processes
• Measurable linear transformations
• Prediction
• Large dimensions

This article deals with linear prediction in large dimensions. One obtains various explicit forms of the best linear predictor in a Hilbert space. The difficulty comes from the fact that the associated linear operator is, in general, not continuous. Applications to ARMAH processes, models with noise and Bayesian estimators are considered.

• 出版日期2014-2