Title

FORECASTING TIME SERIES WITH NEURAL NETS

Date of Award

1999

Degree Type

Thesis

Degree Name

Master of Science

Department

Statistics and Actuarial Sciences

First Advisor

A.I. McLeod

Abstract

FFNN Feed Forward Neural Nets are one of the most widely used neural nets. In this thesis the FFNN architecture is examined and compared with statistical time series models for a variety of time series prediction problems FFNN do not assume any probability models while statistical models are based on the probability model. Therefore, if the goal of the modelling is rigorous quantication of uncertainty statistical models are more suitable. However if the goal is merely prediction we demonstrate that neural nets have a lot to offer Widely different parameter settings in the neural net approach often lead to models which make virtually the same predictions. Neural net offer a more exible approach to model building which is especially helpful in nonlinear and nonGaussian situations. In this thesis the performance by NN models and statistical models for prediction is examined by using visualization techniques and statistical tests.

Comments

at the request of A.I. McLeod

This document is currently not available here.

Share

COinS