Distribution Theory of Runs and Run-Related Statistics in Sequences of Dependent Trials
Loading...
Files
Date
2013
Authors
Yalçın, Femin
Journal Title
Journal ISSN
Volume Title
Publisher
İzmir Ekonomi Üniversitesi
Open Access Color
OpenAIRE Downloads
OpenAIRE Views
Abstract
Tekrarlar ve tekrarlarla ilgili istatistikler, sistem güvenilirliği, istatistiksel süreç kontrolü ve hipotez testi gibi çeşitli uygulamalı alanlarda kullanılmaktadır. Literatürde hem bağımsız hem de bağımlı deneme dizileri üzerinde tanımlanan tekrarların dağılımlarını elde etmek için çeşitli yöntemler geliştirilmiştir. Bu tezde, biri Markov bağımlılık olmak üzere iki farklı bağımlılık türü ele alınmıştır. Diğer bağımlılık modeline göre bir denemenin başarılı ya da başarısız olması, kendisinden önceki denemelerdeki toplam başarılı deneme sayısına bağlıdır. Bazı tekrar istatistiklerinin dağılımları elde edilmiş ve teorik sonuçların uygulamaları açıklayıcı örneklerle pekiştirilmiştir.
Runs and run-related statistics have been successfully used in various applied fifields including system reliability, statistical process control, and hypothesis testing. Various methods have been developed in the literature to investigate both exact and asymptotic distributions of runs not only for independent trials but also for a sequence of dependent trials. In this thesis, two di fferent types of dependence have been considered one of which is the Markovian type of dependence. According to the other dependence model, the outcome of the present trial depends on the total number of successful trials so far. Distributions of some run statistics under these dependence types have been obtained and applications of the theoretical results have been established with illustrative examples.
Runs and run-related statistics have been successfully used in various applied fifields including system reliability, statistical process control, and hypothesis testing. Various methods have been developed in the literature to investigate both exact and asymptotic distributions of runs not only for independent trials but also for a sequence of dependent trials. In this thesis, two di fferent types of dependence have been considered one of which is the Markovian type of dependence. According to the other dependence model, the outcome of the present trial depends on the total number of successful trials so far. Distributions of some run statistics under these dependence types have been obtained and applications of the theoretical results have been established with illustrative examples.
Description
Keywords
Tekrarlar, Bernoulli denemeleri, Dağılımlar, Markov zinciri, Geçmiç toplama dayalı bağımlılık modeli., Runs, Bernoulli trials, Exact distributions, Markov chain, Previoussum dependent model., Matematik, Mathematics, İstatistik, Statistics
Turkish CoHE Thesis Center URL
Fields of Science
Citation
WoS Q
N/A
Scopus Q
N/A
Source
Volume
Issue
Start Page
1
End Page
84
