Research Article
BibTex RIS Cite

Kısıtlandırılmamış Kısmi Oransal Odds Modelinin Doğru Sınıflandırma Performansı Üzerine Bir Çalışma

Year 2017, Volume: 3 Issue: 3, 14 - 26, 27.12.2017

Abstract

Birimlerin iki ya da daha fazla düzeyli kategorik değişkenler bakımından sınıflandırılmasında birçok yöntem kullanılmaktadır. Bu yöntemlerden bazıları, bağımlı değişken düzeyinin ikiden fazla ve sıralı bir yapıda olması durumunda kullanılan oransal odds modeli ve bu modele ait temel varsayımın bazı değişkenler için sağlanıp bazı değişkenler için sağlanmaması durumunda kullanılan kısıtlandırılmamış kısmi oransal odds modelidir. Bu çalışmada kısıtlandırılmamış kısmi oransal odds modeli ele alınarak bağımlı değişken düzeyinin sayısı, bağımsız değişken sayısı ve örneklem büyüklüğü değiştirildiğinde doğru sınıflandırma oranları incelenmiştir. Ayrıca bağımsız değişkenlerin tümünün sürekli olması durumu için simülasyon çalışması yapılmış ve bu veriler için oransal odds modeli, kısıtlandırılmamış kısmi oransal odds modeli, doğrusal diskriminant analizi ve karesel diskriminant analizi yöntemlerinin birbirlerine göre üstünlükleri ortaya konulmaya çalışılmıştır.

References

  • [1] Abdel A. and Wang X. (2008). Analysis of leftturn crash injury severity by conflicting pattern using partial proportional odds models. Department of Civil&Environmental Engineering, University of Central Florida, Orlando, United States.
  • [2] Adeleke K.A. and Adepoju A.A. (2009). Ordinal logistic regression model: an application to pregnancy outcomes journal of mathematics and statistics. International Journal of Epidemilogy, Great Britain. 279-285, 2010 ISSN 1549-364.
  • [3] Ananth C. and Kleinbaum D.G. (1997). Regression models for ordinal responses: A review of methods and applications. International Journal of Epidemilogy, Great Britain.
  • [4] Chen C.K. and Hughes J. (2004). Using ordinal regression model to analyze student satisfaction questionnaires. Association for Institutional Research, Volume1.
  • [5] Dağlıoğlu H.( 2014). KısıtlandırılmamıĢ Kısmi. Oransal Odds Modelinin Doğru Sınıflandırma Performansı Üzerine Bir ÇalıĢma, Gazi Üniversitesi, Fen Bilimleri Enstitüsü, Doktora Tezi, Ankara.
  • [6] Damodar G. (1995). Basic econometrics. Ġstanbul, Third Edition. 1995, pp.541.
  • [7] Fujimoto K. (2005). From women‟s college to work: inter-organizational networks in the Japanese female labor market. Social Science Research 34 (4), 651–681.
  • [8] Lall R., Campbell M. J., Walters S. J., Morgan K. (2002). A review of ordinal regression models applied on health-related quality of life assessments. Statistical Methods in Medical Research 11: 49–67.
  • [9] Liao T. F. (1994). Interpreting probability models: lojit, probit, and other generalized linear models. Sage Publications, Thousand Oaks, CA
  • [10] Long S. J. (1997). Regression models for categorical and limited dependent variables. Sage Publications, Thousand Oaks, CA.
  • [11] McCullagh P. (1980) Regression models for ordinal data. Journal of the Royal Statistical Society. Series B Volume 42, Issue 2, 109-142.
  • [12] McCullagh P. and Nelder, J. A. (1989). Generalized Linear Models, Second Edition, Chapman and Hall, London.
  • [13] Peterson B.L. (1986). Proportional odds and partial proportional odds models for ordinal response variables. Department of Biostatistics, University of North Coralino, at Chapel Hill Institute of statistics mimeo series no, October 1986
  • [14] Peterson B. and Harrell, F. E. (1990). Partial proportional odds models for ordinal response variables. Applied Statistics, 39, 205-217.
  • [15] Williams R. (2005) Gologit2: a program for generalized logistic regression/ partial proportional odds models for ordinal variables. Retrieved May 12, 2005.
  • [16] Williams R. (2006). Generalized ordered lojit/partial proportional odds models for ordinal dependent variables. Stata Journal 6 (1), 58–82

A Study On Correct Classıfıcatıon Performance Of Unconstraınt Partıal Proportıonal Odds Model

Year 2017, Volume: 3 Issue: 3, 14 - 26, 27.12.2017

Abstract

Several methods are used to classify units in terms of categorical variables having two or more levels. Some of these methods are proportional odds model which is used when the level of the dependent variable is more than two and a ordinal pattern and the unconstraint partial proportional odds model which is used when the fundamental assumption of this model is established from the some of the variables and is not satisfied some of the variables. In this study, the correct classification rates are investigated with varying the number of classes of dependent variables, the number of independent variables and the sample size by taking unconstraint partial proportional odds model. A simulation study is performed in the case of all independent variables are continuous and also superiority of proportional odds model, unconstraint partial proportional odds model, linear discriminant analyze and quadratic discriminant analyze to each other are tried to revealed for these data.

References

  • [1] Abdel A. and Wang X. (2008). Analysis of leftturn crash injury severity by conflicting pattern using partial proportional odds models. Department of Civil&Environmental Engineering, University of Central Florida, Orlando, United States.
  • [2] Adeleke K.A. and Adepoju A.A. (2009). Ordinal logistic regression model: an application to pregnancy outcomes journal of mathematics and statistics. International Journal of Epidemilogy, Great Britain. 279-285, 2010 ISSN 1549-364.
  • [3] Ananth C. and Kleinbaum D.G. (1997). Regression models for ordinal responses: A review of methods and applications. International Journal of Epidemilogy, Great Britain.
  • [4] Chen C.K. and Hughes J. (2004). Using ordinal regression model to analyze student satisfaction questionnaires. Association for Institutional Research, Volume1.
  • [5] Dağlıoğlu H.( 2014). KısıtlandırılmamıĢ Kısmi. Oransal Odds Modelinin Doğru Sınıflandırma Performansı Üzerine Bir ÇalıĢma, Gazi Üniversitesi, Fen Bilimleri Enstitüsü, Doktora Tezi, Ankara.
  • [6] Damodar G. (1995). Basic econometrics. Ġstanbul, Third Edition. 1995, pp.541.
  • [7] Fujimoto K. (2005). From women‟s college to work: inter-organizational networks in the Japanese female labor market. Social Science Research 34 (4), 651–681.
  • [8] Lall R., Campbell M. J., Walters S. J., Morgan K. (2002). A review of ordinal regression models applied on health-related quality of life assessments. Statistical Methods in Medical Research 11: 49–67.
  • [9] Liao T. F. (1994). Interpreting probability models: lojit, probit, and other generalized linear models. Sage Publications, Thousand Oaks, CA
  • [10] Long S. J. (1997). Regression models for categorical and limited dependent variables. Sage Publications, Thousand Oaks, CA.
  • [11] McCullagh P. (1980) Regression models for ordinal data. Journal of the Royal Statistical Society. Series B Volume 42, Issue 2, 109-142.
  • [12] McCullagh P. and Nelder, J. A. (1989). Generalized Linear Models, Second Edition, Chapman and Hall, London.
  • [13] Peterson B.L. (1986). Proportional odds and partial proportional odds models for ordinal response variables. Department of Biostatistics, University of North Coralino, at Chapel Hill Institute of statistics mimeo series no, October 1986
  • [14] Peterson B. and Harrell, F. E. (1990). Partial proportional odds models for ordinal response variables. Applied Statistics, 39, 205-217.
  • [15] Williams R. (2005) Gologit2: a program for generalized logistic regression/ partial proportional odds models for ordinal variables. Retrieved May 12, 2005.
  • [16] Williams R. (2006). Generalized ordered lojit/partial proportional odds models for ordinal dependent variables. Stata Journal 6 (1), 58–82
There are 16 citations in total.

Details

Journal Section Research Articles
Authors

Hatice Dağlıoğlu This is me

Semra Oral Erbaş

Publication Date December 27, 2017
Submission Date August 15, 2017
Acceptance Date September 20, 2017
Published in Issue Year 2017 Volume: 3 Issue: 3

Cite

IEEE H. Dağlıoğlu and S. O. Erbaş, “A Study On Correct Classıfıcatıon Performance Of Unconstraınt Partıal Proportıonal Odds Model”, GJES, vol. 3, no. 3, pp. 14–26, 2017.

Gazi Journal of Engineering Sciences (GJES) publishes open access articles under a Creative Commons Attribution 4.0 International License (CC BY). 1366_2000-copia-2.jpg