Dik Uzaklıklar Kareler Toplamına Göre Klasik Doğrusal Regresyon ile Ortogonal Regresyonun Karşılaştırılması

Regresyon analizi, değişkenler arasındaki ilişkiyi inceleme ve modellemede kullanan istatistiksel bir tekniktir. Bu çalışmanın amacı ortogonal regresyon (OR) eşitliğini açık bir şekilde sunmak ve dik uzaklıklar kareleri toplamına göre klasik doğrusal regresyon (KDR) ile ortogonal regresyonu karşılaştırmaktır. Bu amaçla analizler bir örnek üzerinden gösterilmiştir. Araştırma sonucunda ortogonal regresyonun dik uzaklıklar kareler toplamının daha küçük olduğu bulunmuştur. Buradan bağımlı ve bağımsız değişkenler arasındaki doğrusal ilişkiyi, ortogonal regresyon doğrusunun klasik doğrusal regresyon doğrusundan daha iyi temsil ettiği görülmüştür. Bu sonuçlara bağlı olarak basit doğrusal regresyon çalışmalarında OR tekniğinin KDR tekniğinden daha doğru sonuçlar elde etmede kullanılabilecek bir regresyon tekniği olduğu düşünülmektedir.

Comparison of Classical Linear Regression and Orthogonal Regression with Respect to the Sum of Squared Perpendicular Distances

Regression analysis is a statistical technique for investigating and modeling the relationship between variables.The purpose of this study was the trivial presentation of the equation for orthogonal regression (OR) and thecomparison of classical linear regression (CLR) and OR techniques with respect to the sum of squaredperpendicular distances. For that purpose, the analyses were shown by an example. It was found that the sumof squared perpendicular distances of OR is smaller. Thus, it was seen that OR line has appeared to present amuch better fit for the data than CLR line. Depending on those results, the OR is thought to be a regressiontechnique to obtain more accurate results than CLR at simple linear regression studies.

PDF

___

Adcock, R.J. (1878). A Problem in least squares, Annals of Mathematics, 5(2), 53-54.
Akdeniz, F. (2013). Olasılık ve İstatistik, Nobel Kitabevi, 18. Press, Ankara, 459-492.
Alma, Ö.G., & Vupa, Ö. (2008). Regresyon Analizinde kullanılan En Küçük Kareler ve En küçük Medyan Kareler Yöntemlerinin Karşılaştırılması, Süleyman Demirel Üniversitesi Fen Edebiyat Fakültesi Fen Dergisi, 3(2), 219-229.
Atar, B. (2010). Basit Doğrusal Regresyon Analizi ile Hiyerarşik Doğrusal Modeller Analizinin Karşılaştırılması, Eğitimde ve Psikolojide Ölçme ve Değerlendirme Dergisi, 1(2), 78-84.
Bahar, H.H. (2006). An Evaluation of KPSS Scores According to Grade Point Avarege and Gender, Education and Science, 31(140), 68-74.
Bağçeci, B., Döş, B., & Sarıca, R. (2011). An Analysis of Metacognitive Awareness Levels and Academic Achievement of Primary School Students, Mustafa Kemal University Journal of Social Sciences Institute, 8(16), 551-566.
Bayat, N., Şekercioğlu, G., & Bakır, S. (2014). Okuduğunu Anlama ve Fen Başarısı Arasındaki İlişkinin Belirlenmesi, Eğitim ve Bilim, (tedmem), 39(176), 457-466.
Büyüköztürk, Ş., Çokluk, Ö., & Köklü, N. (2012). Sosyal bilimler için İstatistik. Ankara: Pegem Akademi.
Calzada, M.E., & Scariano, S.M. (2003). Contrasting total least squares with ordinary least squares part II: Examples and Comparisons, Mathematics and Computer Education, 37(2), 159-174.
Carr, J.R. (2012). Orthogonal regression: A Teaching perspective, International Journal of Mathematical Education in Science and Technology, 43(1), 134-143.
Carroll, R.J., & Ruppert D. (1996). The use and misuse of orthogonal regression in linear errors-in-variables models, The American Statistician, 50 (1), 1-6.
Coşkuntuncel, O. (2013). The Use of alternative Regression Methods in Social Sciences and the Comparison of Least Squares and M Estimation Methods in Terms of the Determination of Coefficient, Educational Sciences: Theory & Practice, 13(4), 2139-2158.
Deming, E.W. (1943). Statistical Adjustment of Data, Dover Publications, Inc., New York.
Ding, G., Chu, B., Jin, Y., & Zhu, C. (2013). Comparison of orthogonal regression and least squares in measurement error modeling for prediction of material property, Nanotechnology and Material Engineering Research. Advanced Materials Research, 661, 166-170.
Doruk, M., Özdemir, F., & Kaplan, A. (2015). The Relationship Between Prospective Mathematics Teachers’ Conceptions on Constructing Mathematical Proof and Their Self-Efficacy Beliefs Towards Mathematics, Kastamonu Üniversitesi Kastamonu Eğitim Dergisi, 23(2), 861-874.
Elfessi, A., & Hoar, R.H. (2001). Simulation study of a linear relationship between two variables affected by errors, Journal of Statistical Computation and Simulation, 71(1), 29-40.
Fišerová, E., & Hron, K. (2010). Total least squares solution for compositional data using linear models, Journal of Applied Statistics, 37(7), 1137-1152.
Genç, Y., Sertkaya, D., & Demirtaş, S. (2003). Klinik Araştırmalarda İki Ölçüm Tekniğinin Uyumunu İncelemede Kullanılan İstatistiksel Yöntemler. Ankara Üniversitesi Tıp Fakültesi Mecmuası, 56(1), 1- 6.
Glaister, P. (2005). The use of orthogonal distances in generating the total least squares estimate, Mathematics and Computer Education, 39(1), 21-30.
Golub, G.H., & Van Loan, C.F. (1980). An Analysis of the total least squares problem, SIAM Journal on Numerical Analysis, 17(6), 883-893.
İlğan, A., Erdem, M., Yapar, B., Aydın. S., & Aydemir, Ş.Ş. (2012). Parents Interest and Regression Level of Primary State School Students Level Determination Exam (SBS) Score, Journal of Educational Sciences Research, 2(2), 1-17.
Isobe, T., Feigelson, E.D, Akritas, M.G., & Babu, G.J. (1990). Linear regression in astronomy I, The Astrophysical Journal, 364, 104-113.
Kane, M.T., & Mroch, A.A. (2010). Modeling group differences in OLS and Orthogonal Regression: Implications for differential validity studies, Applied Measurement in Education, 23, 215-241.
Kermack, K.A., & Haldane, J.B.S. (1950). Organic correlation and allometry, Biometrika Trust, 37(1), 30–41.
Kesicioğlu, O.S., & Güven, G. (2014). Investigation of the Correlation Between Preservice Early Childhood Teachers' Self-Efficacy Levels and Problem Solving, Empathy and Communication Skills, Turkish Studies, International Periodical For the Languages, Literature and History of Turkish or Turkic, 9(5), 1371-1383.
Lane, D. M. (2016). “Introduction to Linear Regression”, http://onlinestatbook.com/2/regression/intro.html Accessed 21 February 2016.
Leng, L., Zhang, T, Kleinman, L., & Zhu, W. (2007). Ordinary Least Square Regression, Orthogonal Regression, Geometric Mean Regression and Their Applications in Aerosol Science, Journal of Physics, Conference Series 78(1), 1-5, http://iopscience.iop.org/1742-6596/78/1/012084 Accessed 17 March 2016.
Li, H.C. (1984). Generalized problem of least squares, The American Mathematical Monthly, 91(2), 135-137.
Lolli, B., & Gasperini, P. (2012). A comparison among general orthogonal regression methods applied to earthquake magnitude conversions, Geophysical Journal International, 190(2), 1135-1151.
Ludbrook, J. (2010). Linear regression analysis for comparing two measurers or methods of measurement: But which regression?, Clinical and Experimental Pharmacology and Physiology, 37, 692-699.
Markovsky, I., & Van Huffel, S. (2007). Overview of total least- squares methods, Signal Processing, 87, 2283-2302.
Montgomery D.C., Peck E.A., & Vining G.G. (2012). Introduction to Linear Regression Analysis, 5th Edition, 1-11. http://eu.wiley.com/WileyCDA/WileyTitle/productCd-0470542810.html Accessed 11 February 2016.
Nievergelt, Y. (1994). Total least squares: State-of-the-art regression in numerical analysis, SIAM Review (Society for Industrial and Applied Mathematics), 36(2), 258–264.
Ortiz, J.V., Pogliani, L., & Besalú, E. (2010). Two-variable linear regression: Modeling with orthogonal leastsquares analysis, Journal of Chemical Education, 87(9), 994-995.
Öztürk, S. (2012). İstatistiksel Regresyon Yöntemlerinin Farklı Veri Gruplarına Uygulanması Üzerine Bir Analiz, Gümüşhane Üniversitesi Fen Bilimleri Enstitüsü Dergisi, 2(2), 55-67.Pearson, K. (1901). On lines and planes of closest fit to systems of points in space, Philosophical Magazine, 2, 559-572.
Saraçlı, S. (2011). Tip II Regresyon Tekniklerinin Monte-Carlo Simülasyonu ile Karşılaştırılması, e-Journal of New World Sciences Academy, 6(2), 26-35.
Saraçlı, S., & Çelik, H.E. (2012). Metot Karşılaştırma Çalışmalarında Bland-Altman ve Tip II Regresyon Analizinin Karşılaştırılması, Düzce Üniversitesi Sağlık Bilimleri Enstitüsü Dergisi, 2(1), 11-14.
Saraçlı, S., Doğan, İ., & Doğan, N. (2009). Medikal Metod Karşılaştırma Çalışmalarında Deming Regresyon Tekniği, Turkiye Klinikleri J Biostat, 1(1), 9-15.
Scariano, S.M., & Barnet, II.W. (2003). Contrasting total least squares with ordinary least squares part I: Basic ideas and result, Mathematics and Computer Education, 37(2), 141-158.
Sykes, A.O. (1993). An Introduction to Regression Analysis, Coase-Sandor Institute for Law & Economics Working, 20, 1-34.
Stöckl, D., Dewitte, K., & Thienpont, L.M., (1998). Validity of linear regression in method comparison studies: is it limited by the statistical model or the quality of the analytical input data?, Clinical Chemistry, 44(11), 2340–2346.
Üredi, I., & Üredi, L. (2005). İlköğrtim 8. Sınıf Öğrencilerinin Öz-düzenleme Stratejileri ve Motivasyonel İnançlarının Matematik Başarısını Yordama Gücü, Mersin University Journal of the Faculty of Education, 1(2), 250-260.
Warton, D.I., Wright, I.J., Falster, D.S. & Westoby, M. (2006). Bivariate line-fitting methods for allometry, Biological Reviews, 81(2), 259-291.