SavedAdd your journal

Journals / Süleyman Demirel Üniversitesi Fen Bilimleri Enstitüsü Dergisi / 2016 / Cilt: 20 - Sayı: 2

İç İçe Hata Regresyon Modelinde Sabit Etki Parametresi için Güven Aralığı Metodlarının Karşılaştırılması

A Comparisonof Confidence Interval Methods of Fixed Effectin Nested Error Regression Model

HATİCE TÜL KÜBRA AKDUR
DENİZ ÖZONUR
HÜLYA BAYRAK
Pages
167–175
DOI
—
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

It is observed that parametric bootstrap-based method provides bettercoverage rates for small intra-correlation and profile likelihood method usuallyprovides better results for moderate and strong correlation.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:Linear mixed-effects models are very popular and powerful tools inmany scientific fields such as zoology, biology, and education.
📊
Key Findings & Data:Estimators of fixedeffects do not only depend on the variances of error terms but they also depend onrandom terms in mixed-effect models.
💡
Conclusion & Impact:It is observed that parametric bootstrap-based method provides bettercoverage rates for small intra-correlation and profile likelihood method usuallyprovides better results for moderate and strong correlation.
🔬Design:Modeling & Simulation
📍Scope:Comprehensive Sample
⚙️Variables:Domain Parameters

Abstract

Linear mixed-effects models are very popular and powerful tools inmany scientific fields such as zoology, biology, and education. Estimators of fixedeffects do not only depend on the variances of error terms but they also depend onrandom terms in mixed-effect models. When the distributions of random effects are unknown or enough sample size cannot be obtained, standard methods may fail. This study aims to determine a promising confidence interval method amongexisting methods in terms of coverage probability of true value of parameter.Standard and parametric bootstrap-based confidence interval methods for nestederror regression model were compared in the simulation study under smallsamples. It is observed that parametric bootstrap-based method provides bettercoverage rates for small intra-correlation and profile likelihood method usuallyprovides better results for moderate and strong correlation.

Özet

Doğrusal karışık etki modelleri, zooloji, biyoloji ve eğitim gibi bilimin birçokalanında popüler ve güçlü bir araçtır. Bu modellerde sabit etki parametrelerinintahminleri hata ve rasgele etki terimlerinin varyanslarına bağlıdır. Rasgeleetkilerin dağılımları bilinmediğinde veya yeterli sayıda örnek bulunamadığında standart metodlar doğru sonuç vermeyebilir. Karışık etki modellerinin özel bir haliolan iç içe hata regresyon modelinin sabit etki parametresi için varolan güvenaralığı metodları arasında parametrenin gerçek değerini kapsama olasılığı bakımından en iyi olan güven aralığı metodu aranmıştır. Standart ve parametrikbootstrap-tabanlı güven aralığı metodları iç içe hata regresyon modeli içinsimülasyon çalışmasında küçük örnek çaplarında karşılaştırılmıştır. Zayıfkorelasyonda parametrik-bootstrap metodları daha iyi sonuçlar vemiştir. Profilolabilirlik metodu orta ve güçlü korelasyonlarda daha iyi sonuçlar sağlamıştır.

🔖

Cite This Article

AKDUR, H. T. K., ÖZONUR, D., BAYRAK, H. (2016). İç İçe Hata Regresyon Modelinde Sabit Etki Parametresi için Güven Aralığı Metodlarının Karşılaştırılması. Süleyman Demirel Üniversitesi Fen Bilimleri Enstitüsü Dergisi, Cilt: 20 - Sayı: 2, 167–175. https://ajindex.com/articles/310715

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (9)

  1. Demidenko, E. 2004. Mixed Models: Theory and Applications. Wiley, 726p. Eisenhart, C. 1947. The Assumptions Underlying the Analysis of Variance. Biometrics, 3(1), 1-21.
  2. Patterson, H.D., Thompson, R. 1971. Recovery of Inter-block Information When Block Sizes are Unequal. Biometrika, 58(3), 545-554.
  3. Searle, S.R., Casella, G., McCulloch, C.E., 2009. Variance Components (Vol. 391). John Wiley & Sons, 509p. Kackar, R. N., Harville, D. A. 1984. Approximations for Standard Errors of Estimators of Fixed and Random Effects in Mixed Linear Models. Journal of the American Statistical Association, 79(388), 853-862.
  4. Kenward, M. G. Roger, J. H. 1997. Small Sample Inference for Fixed Effects From Restricted Maximum Likelihood. Biometrics, 1997, 983- 997. Harville, D. A., Fenech, A. P. 1985. Confidence Intervals for a Variance Ratio, or for Heritability, in Model. Biometrics, 1985, 137-152.
  5. Mixed Linear Savin, A., 2005. Confidence Intervals for Common Mean in One-way Classification Model with Fixed Effects. Measurement, 2005, 15-19.
  6. Hall, P., Maiti, T. 2006. On Parametric Bootstrap Methods for Small Area Prediction. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 68(2), 221-238.
  7. Staggs, V. 2009. Parametric Bootstrap Interval Approach to Inference for Fixed Effects in the Mixed Linear Model, PhD Thesis, University of Kansas, 73 p, Lawrence. Burch, B. D. 2011. Confidence Intervals for Variance Components in Unbalanced One-way Random Effects Model Using Non-normal Distributions. Journal of Statistical Planning and Inference, 141(12), 3793-3807.
  8. Liu, J. 2013. Statistical Inference for Functions of the Parameters of a Linear Mixed Model, PhD Thesis, Iowa State University, 75p, Ames. Neyman, J., Pearson, E. S. 1928. On the Use and Interpretation of Certain Test Criteria for Purposes I. Biometrika, 1928, 175-240. Inference: Part Wald, A. 1943. Tests of Statistical Hypotheses Concerning Several Parameters when the Number of Observations is Large. Transactions ofthe American Mathematical Society, 54(3), 426-482.
  9. Bates, D. M., Pinheiro, J. C. 1994. Model Building for Nonlinear Mixed Effects Models, University of Wisconsin, Department of Biostatistics. Lehmann, E. L. 1999. Elements of Large-Sample Theory. Springer Science & Business Media, 645p. Hulting, F. L., Harville, D. A. 1991. Some Bayesian and Non-Bayesian Procedures for the Analysis of Comparative Experiments and for Small-area Estimation: Computational Aspects, Frequentist Properties, and Relat