Öz
Çoklu lineer regresyon modellerinde çoklu iç ilişki olması durumunda, EKK tahmin edicisi yansızlığını korur, ancak varyansı büyür ve mutlak değerce büyük parametre tahminleri vermektedir. Bu nedenle X X matrisinin kötü koşulluluğunu ortadan kaldırmaya ve varyansı küçültmeye yönelik çeşitli yanlı tahmin ediciler geliştirilmiş ve önerilmiştir. r-k-d sınıf tahmin edici bunlardan biridir. Bu tahmin edici 7 farklı tahmin ediciyi özel durumlar olarak ihtiva ettiğinden oldukça esnek ve geniş kapsamlı bir tahmin edicidir. Bu çalışmada, r-k-d sınıf tahmin edicinin ampirik çalışmalarda ne denli kullanışlı olduğu gösterilmiştir.
Anahtar Kelimeler
Çoklu İç İlişki r-k-d Sınıf Tahmin Edici Tahmin Kestirim PRESS
Kaynakça
- BAYE, M. R., and PARKER, D. F., 1984. Combining Ridge and Principal Component Regression: A Money Demand Illustration. Comm. Statist. Theory Methods, 13, 2, 197-205.
- HOERL, A. E., and KENNARD, R. W., 1970. Ridge Regression: Biased Estimation for Nonorthogonal Problems. Technometrics, 12, 1, 55-67.
- İNAN, D., 2011. Combining the Liu-Type Estimator and the Principal Component Regression Estimator: A Comparative Simulation Study. European Young Statisticians Meeting, 17th EYMS. Lisbon, Portugal.
- KAÇIRANLAR, S., and SAKALLIOĞLU, S., 2001. Combining the Liu Estimator and the Principal Component Regression Estimator. Comm. Statist. Theory Methods, 30, 12, 2699-2705.
- LIU, K., 1993. A New Class of Biased Estimate in Linear Regression. Comm. Statist. Theory Methods, 22, 2, 393-402.
- LIU, K., 2003. Using Liu-Type Estimator to Combat Collinearity. Comm. Statist. Theory Methods, 32, 5, 1009-1020.
- MONTGOMERY, D. C., and FRIEDMAN D. J., 1993. Prediction Using Regression Models with Multicollinear Predictor Variables. IIE Transactions, 25 (3), 73- 85.
- STEIN C. 1956. Inadmissibility of usual estimator for the mean of a multivariate normal distribution. Proceedings of the Third Berkeley Symposium on Mathematical Statistics and Probability, 1: 197-206.
Öz
In the presence of multicollinearity the OLS estimator remains unbiased, but the variances and the absolute values of the parameter estimations are tend to be large. Various biased estimators are developed and proposed in order to eliminate ill conditioning of X’X matrix and reduce the variance. One of them is r-k-d class estimator. This estimator is quite flexible and extensive because it contains 7 different estimators as special cases. The main aim of this paper is to show how usable this estimator is in empirical studies.
Anahtar Kelimeler
Multicollinearity r-k-d class estimator estimation prediction PRESS
Kaynakça
- BAYE, M. R., and PARKER, D. F., 1984. Combining Ridge and Principal Component Regression: A Money Demand Illustration. Comm. Statist. Theory Methods, 13, 2, 197-205.
- HOERL, A. E., and KENNARD, R. W., 1970. Ridge Regression: Biased Estimation for Nonorthogonal Problems. Technometrics, 12, 1, 55-67.
- İNAN, D., 2011. Combining the Liu-Type Estimator and the Principal Component Regression Estimator: A Comparative Simulation Study. European Young Statisticians Meeting, 17th EYMS. Lisbon, Portugal.
- KAÇIRANLAR, S., and SAKALLIOĞLU, S., 2001. Combining the Liu Estimator and the Principal Component Regression Estimator. Comm. Statist. Theory Methods, 30, 12, 2699-2705.
- LIU, K., 1993. A New Class of Biased Estimate in Linear Regression. Comm. Statist. Theory Methods, 22, 2, 393-402.
- LIU, K., 2003. Using Liu-Type Estimator to Combat Collinearity. Comm. Statist. Theory Methods, 32, 5, 1009-1020.
- MONTGOMERY, D. C., and FRIEDMAN D. J., 1993. Prediction Using Regression Models with Multicollinear Predictor Variables. IIE Transactions, 25 (3), 73- 85.
- STEIN C. 1956. Inadmissibility of usual estimator for the mean of a multivariate normal distribution. Proceedings of the Third Berkeley Symposium on Mathematical Statistics and Probability, 1: 197-206.
Ayrıntılar
| Birincil Dil | Türkçe |
|---|---|
| Bölüm | Araştırma Makaleleri |
| Yazarlar | |
| Yayımlanma Tarihi | 1 Haziran 2013 |
| Gönderilme Tarihi | 11 Ağustos 2015 |
| Yayımlandığı Sayı | Yıl 2013 Cilt: 17 Sayı: 1 |