Estimating total number of unemployment using ARMA model

Tay, Sze Hui (2010) Estimating total number of unemployment using ARMA model. Universiti Malaysia Sabah. (Unpublished)


Download (8MB) | Preview


Data that has been used to forecast in this study is the total number of unemployment in Malaysia. The trend and changes of monthly total unemployment are identified and observed. The total number of unemployment from January 1988 to June 2009 is used for data analysis. In the meanwhile, the total number of unemployment from July 2009 to December 2009 is used for accuracy checking. The data set is analyzed by using Autoregression Moving Average (ARMA) method which is compared to the Box-Jenkins approach. Both of these two best models are voted out for the "best of the best" model via eight selection criteria (8SC) by determining the lowest values. The best model determined by emphasizing the elimination of the lags with the most insignificant coefficient is superior to alternative best model suggested by Box-Jenkins approach. Furthermore, mean absolute percentage error (MAPE) examines the consistency and precision of the forecasting technique used. The forecast result shows that the ARMA model is only suitable to forecast in a short term of period.

Item Type: Academic Exercise
Uncontrolled Keywords: total unemployment, ARMA Model, box-jenkins approach
Subjects: Q Science > QA Mathematics
Divisions: SCHOOL > School of Science and Technology
Depositing User: ADMIN ADMIN
Date Deposited: 02 Jun 2016 00:37
Last Modified: 30 Oct 2017 03:24

Actions (login required)

View Item View Item