Research Reports from the Department of Operations

Document Type

Thesis

Publication Date

5-1-1984

Abstract

A forecasting methodology has been developed which is based on information theory. This methodology uses the idea that autoregressive (AR) representation of a stationary, Normally distributed time series is equivalent to a model of such a time series which has maximum information content. To apply this idea, two programs were developed. The first program utilizes an Adaptive Sequential Segmentation Algorithm, which is based on an information theoretic measure, to detect subseries in a piecewise stationary time series. This is done to accommodate abrupt or discontinuous changes in the underlying process parameters. A second program utilizes the Akaike Information Criterion to determine the optimal degree of differencing and optimal AR order of an AR model fitted to a given series or subseries. Forecasts and the frequency spectrum of this model are computed. Performance tests of this methodology, and a comparison with the Box-Jenkins methodology is included.

Keywords

Operations research, Time-series analysis, Forecasting--Statistical methods, Information theory, Autoregression (Statistics), Stochastic processes, Algorithms, Akaike information criterion

Publication Title

Master's thesis/Technical Memorandums from the Department of Operations, School of Management, Case Western Reserve University

Issue

Technical memorandum no. 534 ; Submitted in partial fulfillment of the requirements for the Degree of Master of Science.

Rights

This work is in the public domain and may be freely downloaded for personal or academic use

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.