A class of multi-scale time series models.pdfVIP

A Class of Multi-Scale Time Series Models Marco A. R. Ferreira 1,2 , Mike West 1 , Herbert K. H. Lee 1 and David Higdon 1,3 1 ISDS - Duke University, 2 DME - UFRJ, 3 LANL. Summary. We introduce a class of multi-scale models for time series. The novel framework couples ’simple’ standard Markov models for the time series stochastic process at different levels of aggregation, and links them via ’error’ models to induce a new and rich class of structured linear models reconciling modeling and information at different levels of resolution. Jeffrey’s rule of conditioning is used to revise the implied distributions and ensure that the probability distributions at different levels are strictly compatible. Our construction has several interesting characteristics: a variety of autocorrelation functions resulting from just a few pa- rameters; the ability to combine information from different scales; and the capacity to emulate long memory processes. There are at least three uses for our multi-scale framework: to in- tegrate the information from data observed at different scales; to induce a particular process when the data is observed only at the finest scale; as a prior for an underlying multi-scale process. Bayesian estimation based on MCMC analysis and forecasting based on simulation are developed. Two interesting applications are presented: in the first application, we illustrate some basic concepts of our multi-scale class of models through the analysis of a series of potential hydroelectric energy; In the second application we use our multi-scale framework to model a series of land temperatures of the northern hemisphere. 1. Introduction In this paper, we introduce a class of multi-scale models for time series. The construction of these multi-scale models is somewhat subtle: We start with an autoregressive or ARMA process for each level, each level is connected with the immediately finer level through a linear function plus Gaussian noise, and then we use Jeffrey’s rule