Title

Forecasting seasonal time series using fuzzy methods based on the SARIMA model

Date of Publication

2014

Document Type

Bachelor's Thesis

Degree Name

Bachelor of Science in Mathematics

Subject Categories

Physical Sciences and Mathematics

College

College of Science

Department/Unit

Mathematics and Statistics Department

Thesis Adviser

Paolo Lorenzo Y. Bautista

Defense Panel Chair

Frumencio F. Co

Defense Panel Member

Robert Neil F. Leong
Christopher Thomas R. Cruz

Abstract/Summary

Fuzzy time series is a useful alternative to conventional time series methods especially when there is uncertainty in the data. Further developments in the method have been created ever since its introduction in 1993. Although fuzzy time series is slowly getting recognized and more accepted as an alternative to crisp time series, few studies focus on data that have seasonality in them. Seasonal time series is present in stock markets, meteorology, agriculture, and more areas concerned with economics and nature, thus being frequently encountered in practice. There have been different methods of fuzzy time series in forecasting with seasonality. This paper focuses on developing a model guided by a seasonal ARIMA model, clustering the observations through fuzzy c-means, and determining fuzzy relationship using artificial neural networks. The method is compared with the performance of the SARIMA model.

Abstract Format

html

Language

English

Format

Electronic

Accession Number

CDTU019198

Shelf Location

Archives, The Learning Commons, 12F, Henry Sy Sr. Hall

This document is currently not available here.

Share

COinS