Multilevel Modelling with Eigenvector Spatial Filtering and its Application to UN Score Data in Kendari

Authors

  • Lemma Firari Boer Graduate Student in Department Statistics, Bogor Agricultural University, Jl. Meranti Wing 22 level 4 Kampus IPB Darmaga, Bogor (16680), Indonesia
  • Hari Wijayanto Lecturer in Department Statistics, Bogor Agricultural University, Jl. Meranti Wing 22 level 4 Kampus IPB Darmaga, Bogor (16680), Indonesia
  • Indahwati Indahwati Lecturer in Department Statistics, Bogor Agricultural University, Jl. Meranti Wing 22 level 4 Kampus IPB Darmaga, Bogor (16680), Indonesia

Keywords:

Spatial dependence, Eigenvector Spatial Filtering, Multilevel Model.

Abstract

Spatial dependence is a condition where locations will affect its neighborhood that tend to have the same characteristics or attributes. Eigenvector spatial filtering (ESF) is a method initially used to overcome spatial dependence in one-level linear regression by adding the eigenvector function that describes spatial effect from model. This research aims to combine ESF with multilevel modelling and applied them into the data that has both spatial dependence and hierarchy effect and compare the results with those of conventional multilevel model (without ESF). The results indicate that ESF method gave a smaller variance of level-2 random effect and AIC value. It also can be shown that the students-teachers ratio is the only significant predictor that affect UN score in Kendari at the alpha level of 5%.

References

Anselin L, “Spatial Econometrics: Methods and Models”, Kluwer Academic Publisher, Dordrecht, 1988, pp. 101-103.

Chun Y, Griffith DA, “A quality assessment of eigenvector spatial filtering based parameter estimates for the normal probability model”, Spatial Statistics, Vol. 10, pp. 1-11, Apr. 2014.

Corrado L, Fingleton B, “Multilevel Modelling with Spatial Effect”, University of Strathclyde press, Glasgow, 2011, pp. 1-21.

Goodchild MF, “Geographical Information Science”, Geographical Information System, Vol. 6, pp. 31-45, Feb 1992.

Griffith AD, “Spatial Autocorrelation and Spatial Filtering: Gaining Understanding Through Theory and Scientific Visualization”, Springer, New York, 2003, pp 31-130.

LeSage JP, “The Theory and Practice of Spatial Econometrics”, Department of Economics University of Toledo, Toledo, 1999, pp. 1-20 .

Park YM, KIM Y, “A Spatially Filtered Multilevel Model to Account for Spatial Dependency: Application to Self-Rated Health Status in South Korea”, International Journal of Health Geographics, Vol. 13, pp. 1-6, Feb 2014.

Pierawan AC, Tampubolon G, “Spatial Dependence Multilevel Model of Well-Being Across Region in Europe”, Applied Geography, Vol. 47, pp. 168-176, Feb 2014.

Snijders TAB, Bosker RJ, “Multilevel Analysis: An introduction to basic and advance multilevel modeling”, Sage Publication, London, 1999, pp 38-83.

Tiefelsdorf M, Grifith DA, “Semiparametric filtering of spatial autocorrelation: the eigenvector approach”. Environment and Planning, Vol. 39, pp. 1193-1221, Mar 2007.

Xu H, “Compare Spatial and Multilevel Regression Models for Binary Outcome in Neighborhood Study”,Sociological Methodology, Vol 44, pp. 229-272, Aug 2014.

Downloads

Published

2018-05-03

How to Cite

Boer, L. F., Wijayanto, H., & Indahwati, I. (2018). Multilevel Modelling with Eigenvector Spatial Filtering and its Application to UN Score Data in Kendari. International Journal of Sciences: Basic and Applied Research (IJSBAR), 38(2), 24–33. Retrieved from https://www.gssrr.org/index.php/JournalOfBasicAndApplied/article/view/8808

Issue

Section

Articles