Ortalama Karesel Hata Matrisi Kriterine göre Yanlı Tahmin Ediciler Üzerine Çalışma

Yıl 2018, Cilt: 30 Sayı: 1, 18 - 25, 31.03.2018

Nilgün Yıldız

https://doi.org/10.7240/marufbd.371450

Öz

Çoklu doğrusallığa sahip regresyon modelleri, r- (k, d)
sınıfı tahmincileri, ana bileşen regresyonu, Liu tipi tahminciler gibi çeşitli
tahmincileri kullanarak ele alınabilir. Bu çalışmada, r- (k, d) sınıfı kestiricisinin,
ana bileşenlerin regresyonu, Liu tipi tahmincileri ve sıradan en küçük kareler
üzerinde, ortalama karesel hata matrisi (MSEM) kriteri açısından üstün olduğu
koşulları belirledik. Son olarak, sayısal bir örnek ve Monte Carlo simülasyonu
ile teorik sonuçları gösterdik.

Anahtar Kelimeler

Liu-tipi tahmin edici, r-(k, d )Sınıf Tahmin Edici Temel bileşenler regresyonu, ortalama karesel hata matrisi

The Research on Biased Estimators Based on Mean Square Error Matrix Criteria

Öz

Regression models with
multicollinearity can be tackled by using various estimators such as   class estimators, principal
components regression, Liu-type estimators. In this study, we defined
conditions where the   class estimator is superior over
the biased estimators in terms of mean square error matrix (MSEM) criterion.
Finally, we showed theoretical results by means of a numerical example and a
simulation study.

Anahtar Kelimeler

r-(k, d) class estimator, Liu-type estimators, (r-d) class estimator, Ordinary Ridge Regression Principal Component Regression, Mean square error matrix, Multicollinearity

Kaynakça

• [1] Stein, C., Inadmissibility of the usual estimator for mean of multivariate normal distribution. In Neyman J (ed). Proceedings of the third Berkley symposium on mathematical and statistics probability 1, (1956), 197–206. [2] Inan, D., Combining the Liu-type estimator and the principal component regression estimator. Statistical Papers 56, (2015), 147-156. [3] Rao, C. and Toutenburg, H., Linear Models: Least Squares and Alternatives. New York:Springer-Verlag Inc. 1995. [4] Trenkler, G., Mean Square Error Matrix Comparisons of Estimators in Linear Regression. Communications in Statistics Theory and Methods A.14,( 1985) 2495–2509. [5] Trenkler, G. and Toutenburg, H., Mean squared error matrix comparisons between biased estimators an overview of recent results, Statistical Paper 31 (1990 )165–179. [6] Baksalary, J.K. and Trenkler, G., Nonnegative and positive definiteness of matrices modified by two matrices of rank one, Linear Algebra and Its Application 151, (1991), 169–184. [7] Graybill, F.A., Theory and Application of the Linear Model. Duxbury Press, North Scituate, Massachusetts,1976. [8] Woods, H., Steinour , H.H. and Starke. H.R., Effect of composition of Portland cement on heat evolved during hardening, Industrial and Engineering Chemistry 24, (1932),1207–1214. [9] Kaçıranlar, S., Sakallioglu, S., Akdeniz, F., Styan, G.P.H. and Werner. H.J., A new biased estimator in linear regression and a detailed analysis of the widely-analysed dataset on Portland Cement, Sankhya Indian J Stat 61(B),( 1999), 443–459. [10] Liu, K., Using Liu-type estimator to combat collinearity, Communications in Statistics Theory and Methods 32. (5), (2003),1009–1020. [11] Kibria. B.M.G., Performance of some new ridge regression estimators, Communications in Statistics - Simulation and Computation. 32, (2003),2389-2413.