Multivariate Analysis Adaptive Regression Splines (MARS) on Prediction The Underdeveloped District in 2014
DOI:
https://doi.org/10.34123/jurnalasks.v7i2.26Keywords:
Multivariate Adaptive Regression Splines (MARS), Underdeveloped regency, ClassificationAbstract
The purposes of this research are to build underdeveloped regency model and make a prediction in 2014 based on economic categories, Human Resources (HR), infrastructures, fiscal capacity, accessibility, and regional characteristics with MARS method. MARS is a classification method which can handle highdimensional data with unknown pattern in advance, and can be applied to see the interaction between variables. MARS is an alternative method when the data doesn’t fulfil the parametric statistics assumptions. From MARS model, there are three variables that affect underdeveloped regency, they are consumption expenditure per capita, life expectancy, and percentage of household electricity users. The accuracy of MARS model is very high, 97.83 percent and can be used to make a prediction. Based on MARS model, at the end of the National Development Plan 2010-2014 is predicted a significant transitions in regency’s status. This model can also be used to predict the condition of new regency based on empirical data, because in the earlier classification, the status of regency just follows the status of parent region.
Downloads
References
Budiantara, I.N., Suryadi, F., Otok, B.W., Guritno, S. 2006. Pemodelan BSpline dan MARS pada Nilai Ujian Masuk terhadap IPK Mahasiswa Jurusan Disain Komunikasi Visual UK. Petra Surabaya. Jurnal Teknik Industri Vol 8, No 1, hal 1-13.
Chang, Li-Yen. 2014. Analysis of Bilateral Air Passenger Flows: A NonParametric Multivariate Adaptive Regression Spline Approach.Journal of Air Transport Management 34 : 123-130
Direktoral Jenderal Perimbangan Keuangan Kementerian Keuangan. 2013. Affirmative Policy Dalam Percepatan Pembangunan Daerah Untuk Peningkatan Kesejahteraan Rakyat. Jakarta : Kementerian Keuangan.
Fernandez, J. R. A., Nieto, P. J. G., Muniz, C. D., Anton, J. C. A. 2014. Modelling Eutrophication and Risk Prevention in a Reservoir in the Northwest of Spain by Using Multivariate Adaptive Regression Splines Analysis. Ecological Engineering 68 : 80-89
Friedman, J. H. 1991. Multivariate Adaptive Regression Splines. The Annals of Statistics, Vol. 19, No. 1, hal. 1-141
Hair, J.F, Rolph E. Anderson, Ronald L. Tatham, William C. Black. 2006. Multivariate Data Analysis. Sixth Edition, Pearson Education Prentice Hall, Inc.
Kementerian Pembangunan Daerah Tertinggal. 2010. Rencana Strategis Tahun 2010-2014. Jakarta : KPDT
Nash, M. S. dan David F.B. 2001. Parametric and Non Parametric Logistic Regression for Prediction of Precense/ Absence of an Amphibian. Las Vegas, Nevada : US Environmental Protection Agency Office of Research and Development National Exposure Research Laboratory Environmental Sciences Division
Otok, B. W. 2003. Perbandingan MARS dengan Regresi Logistik pada Respon Biner. Prosiding Seminar Nasional Matematika dan Statistika VI. ITS, Surabaya.
Otok, B. W., Akbar, M. S., Guritno, S., Subanar. 2007. Pendekatan Bootstrap pada Klasifikasi Pemodelan Respon Ordinal. Jurnal Ilmu Dasar, Vol. 8 No. I, hal. 54-67.
Otok, B. W. 2009. Konsistensi dan Asimtotik Normalitas Model Multivariate Adaptive Regression Splines (MARS) pada Respon Biner. Jurnal Ilmu Dasar, Vol. 10 No. 2, hal. 133-140.
Quiros, E., Felicimo, A. M., Cuartero, A. 2009. Testing Multivariate Adaptive Regression Splines (MARS) as a Method of Land Cover Classification of TERRA-ASTER Satellite Images. Sensors 2009, 9.
