Publication: Model Seçiminde Bayesci Yaklaşımların Kullanımı
Loading...
Date
Authors
Journal Title
Journal ISSN
Volume Title
Abstract
Hipotez testi, farklılıklara sahip olan iki temel istatistiksel düsünme tarzı tarafından önerilen çözümlere sahip bir model seçim problemidir. Bu istatistiksel yaklasımlar; Klasik ve Bayesci yaklasımlardır. Bu farklılıkların en önemlisi Bayesci yaklasımdaki önsel seçimi oldugu bilinir. Oysa, gerçekte iki yaklasım arasında temel baska ayrılıkların oldugu da bilinen bir gerçektir. Bu tezde, Bayesci hipotez testlerinin temel basit yönleri özet olarak verilerek, simüle edilen veri üzerine örnekler verilmistir. Standart istatistiksel metotlar model belirsizligini ihmal eder. Veri analizcileri olası model sınıfından bir model seçer ve sanki seçilen model veriyi üretmis gibi isleme devam eder. Bu yaklasım model seçiminde belirsizligi ihmal ederek istatistiksel çıkarımlar için güven aralıklarını daha genis tutar ve daha riskli kararlara neden olur. Oysa Bayesci model ortalaması (BMA) bu model belirsizligini göz önüne alan bir yapı sunar. Bu çalısmada BMA yaklasımını sunarak gerçek hayattan bir probleme uygulaması verilmistir. Uygulamada, BMA örnek kestirim performansını gelistirmistir. Anahtar kelimeler: Bayesci yaklasım, Bayesci hipotez testi,Bayes Faktörü,Bayesci Model ortalaması, Model belirsizligi
Hypothesis testing is a model selection problem for which the solution proposed by the two main statistical streams of thought, frequentists and Bayesians, substantially differ. One may think that this fact might be due to the prior chosen in the Bayesian analysis. However, the Bayesian robustness viewpoint has shown that, in general, this is not so and hence a profound disagreement between both approaches exists. In this thesis, we briefly revise the basic aspects of hypothesis testing for Bayesian procedures and discuss illustrations on simulated data. Standard statistical practice ignores model uncertainty. Data analysts typically select a model from some class of models and then proceed as if the selected model had generated the data. This approach ignores the uncertainty in model selection, leading to over-confident inferences and decisions that are more risky than one thinks they are Bayesian model averaging (BMA) provides a coherent mechanism for accounting for this model uncertainty.. In this study, we discuss BMA approach and present a real life application. In this application, BMA provides improved out-of sample predictive performance. Key words: Bayesian approach, Bayesian Hypotesis testing, Bayesian model averaging, model uncertainty,
Hypothesis testing is a model selection problem for which the solution proposed by the two main statistical streams of thought, frequentists and Bayesians, substantially differ. One may think that this fact might be due to the prior chosen in the Bayesian analysis. However, the Bayesian robustness viewpoint has shown that, in general, this is not so and hence a profound disagreement between both approaches exists. In this thesis, we briefly revise the basic aspects of hypothesis testing for Bayesian procedures and discuss illustrations on simulated data. Standard statistical practice ignores model uncertainty. Data analysts typically select a model from some class of models and then proceed as if the selected model had generated the data. This approach ignores the uncertainty in model selection, leading to over-confident inferences and decisions that are more risky than one thinks they are Bayesian model averaging (BMA) provides a coherent mechanism for accounting for this model uncertainty.. In this study, we discuss BMA approach and present a real life application. In this application, BMA provides improved out-of sample predictive performance. Key words: Bayesian approach, Bayesian Hypotesis testing, Bayesian model averaging, model uncertainty,
Description
Tez (yüksek lisans) -- Ondokuz Mayıs Üniversitesi, 2007
Libra Kayıt No: 12663
Libra Kayıt No: 12663
Keywords
Citation
WoS Q
Scopus Q
Source
Volume
Issue
Start Page
End Page
49
