UUM ETD | Universiti Utara Malaysian Electronic Theses and Dissertation
FAQs | Feedback | Search Tips | Sitemap

Study of Muslims in marital system using markov chain simple exponential smoothing (MCses) technique

Fadhilah, Jamaluddin (2016) Study of Muslims in marital system using markov chain simple exponential smoothing (MCses) technique. Masters thesis, Universiti Utara Malaysia.

[img] Text
s816704_01.pdf
Restricted to Registered users only

Download (8MB)
[img]
Preview
Text
s816704_02.pdf

Download (1MB) | Preview

Abstract

The Markov Chain (MC) model is a popular mathematical model used to observe the flow of data in a system. It can also be used to forecast future values for short-term period. However, most previous studies do not focus on the accuracy of the forecast values. Integration of the MC model and Simple Moving Average (SMA) technique is known to produce higher forecast accuracy than the classical MC model for the case of long-term projection with known previous data. However, Simple Exponential Smoothing (SES) technique is more flexible than SMA because it uses a smoothing constant. Therefore, this study develops modeling steps for MC model in the case of limited data and short-term projection by integrating MC model with SES (MCsEs). The MCsEs hybrid model is used to enhance the MC model and improve the accuracy of the forecast values. Four error measures used to determine the accuracy of this model are mean absolute deviation, mean absolute percentage deviation, mean absolute percentage error and mean square error. This study uses a sample of 6061 Muslim couples data in Pendang, Kedah who are in the marital system for the year 2013 and 2014. The number of Muslims in subsequence year according to gender and age categories is forecasted using proposed MCsEs hybrid model. Comparison with MC and MCsMA models indicates that the developed MCsEs hybrid model has better forecast accuracy. Therefore, the MCsEs hybrid model is the most appropriate model to forecast the number of Muslims in the marital system according to gender and age categories for the year 2014. This model can be used for short-term projection in cases with limited data and is applicable in various fields.

Item Type: Thesis (Masters)
Uncontrolled Keywords: Hybrid markov chain model, Forecast accuracy, Simple exponential smoothing
Subjects: Q Science > QA Mathematics > QA273-280 Probabilities. Mathematical statistics
Divisions: Awang Had Salleh Graduate School of Arts & Sciences
Depositing User: Mr. Badrulsaman Hamid
Date Deposited: 17 Jan 2019 05:55
Last Modified: 17 Jan 2019 05:55
URI: http://etd.uum.edu.my/id/eprint/6988

Actions (login required)

View Item View Item