Proving the Efficiency of Alternative Linear Regression Model Based on Mean Square Error (MSE) and Average Width using Aquaculture Data
Mohamad Arif Awang Nawi1, Wan Muhamad Amir W Ahmad2, Mohamad Shafiq Mohd Ibrahim3, Mustafa Mamat4, Mohd Fadhli Khamis5, Mohamad Afendee Mohamed6
1Mohamad Arif Awang Nawi, School of Dental Science, Universiti Sains Malaysia, Health Campus, Kubang Kerian, Kelantan, Malaysia.
2Wan Muhamad Amir W Ahmad, School of Dental Science, Universiti Sains Malaysia, Health Campus, Kubang Kerian, Kelantan, Malaysia.
3Mohamad Shafiq Mohd Ibrahim, Kulliyah of Dentistry, International Islamic University Malaysia, Kuantan Campus, Pahang, Malaysia.
4Mustafa Mamat, Faculty of Informatics and Computing, Universiti Sultan Zainal Abidin, Terengganu, Malaysia.
5Mohd Fadhli Khamis, School of Dental Science, Universiti Sains Malaysia, Health Campus, Kubang Kerian, Kelantan, Malaysia.
6Mohamad Afendee Mohamed, Faculty of Informatics and Computing, Universiti Sultan Zainal Abidin, Terengganu, Malaysia.
Manuscript received on 18 July 2019 | Revised Manuscript received on 03 August 2019 | Manuscript Published on 10 August 2019 | PP: 377-381 | Volume-8 Issue-2S3 July 2019 | Retrieval Number: B10650782S319/2019©BEIESP | DOI: 10.35940/ijrte.B1065.0782S319
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Multiple linear regressions (MLR) model is an important tool for investigating relationships between several response variables and some predictor variables. This method is very powerful and commonly used in finance, economic, medical, agriculture and many more. The main objective of this paper is to compare mean square error (MSE) and the average width between alternative linear regression models and linear regression model. The alternative method in this study is a combination of four methods, namely multiple linear regression method, the bootstrap method, a robust regression method and fuzzy regression through the construction of algorithms by using SAS software. Typically, the alternative method optimized by minimizing the mean square error (MSE) and average width. The results of the study showed a positive improvement for the estimation of parameters generated through these alternative methods.
Keywords: Alternative Linear Regression, Average Width, Mean Square Error.
Scope of the Article: Data Mining and Warehousing