MULTIPLE MONTE CARLO SIMULATED HIDDEN MARKOV MODEL FOR FUZZY TIME SERIES FORECASTING

Authors

  • A. T. Salawudeen Department of Electrical and Electronic Engineering, University of Jos
  • A. Nurudeen Department of Computer of Electrical Engineering, Ahmadu Bello University, Zaria
  • S. Garba Department of Electrical and Electronic Engineering, University of Jos
  • S. U. Hussein Department of Electrical and Electronic Engineering, Nile University, Abuja
  • M. L. Imam Department of Electrical and Electronic Engineering, University of Jos
  • B. Yahaya Department of Electrical and Electronic Engineering, University of Jos
  • I. F. Egbujo Department of Electrical and Electronic Engineering, University of Jos

Keywords:

FTS HMM Monte Carlo Simulation GA RMSE

Abstract

This paper presents a Monte Carlo based Hidden Markov Model (HMM) for fuzzy time
series forecasting. To make the nature of conjecture and randomness of forecasting more
realistic, the Monte Carlo method with different simulation size is adopted to estimate the
forecasting outcome. To address the insufficiency in data associated with the HMM model,
we adopted a method called smoothing. A number of simulations was performed using
MATLAB simulation environment. The performance of the model was evaluated using the
daily average temperature and cloud density of Taipei, Taiwan. In addition to improving
forecasting accuracy, the proposed model adheres to the central limit theorem, and thus,
the result statistically approximates to the real mean of the target value being forecasted.
Results showed that the proposed model attain and MSE, RMSE, and AFEP of 0.8596,
2.4283, 0.9272 respectively.

Downloads

Published

2019-09-25

Issue

Section

Articles