پديد آورنده :
فاضلي اصل، فروغ
عنوان :
آزمون فرض هاي بيز ناپارامتري
مقطع تحصيلي :
كارشناسي ارشد
گرايش تحصيلي :
آمار رياضي
محل تحصيل :
اصفهان: دانشگاه صنعتي اصفهان، دانشكده علوم رياضي
صفحه شمار :
چهارده، [106]ص.: مصور، جدول، نمودار
يادداشت :
ص.ع. به فارسي و انگليسي
استاد راهنما :
زهرا صابري
استاد مشاور :
افشين پرورده
واژه نامه :
واژه نامه انگليسي به فارسي
توصيفگر ها :
آزمون فرض هاي دو نمونه اي , بيز ناپارامتري , فرايند ديريكله , درخت پوليا
استاد داور :
مجتبي گنجعلي، صفيه محمودي
تاريخ ورود اطلاعات :
1396/07/03
چكيده انگليسي :
Bayesian Nonparametric Hypothesis Testing Forough Fazeli Asl forough fazeli@math iut ac ir 2017 Department of Mathematical Sciences Isfahan University of Technology Isfahan 84156 83111 Iran Supervisor Dr Zahra Saberi z saberi@cc iut ac ir Advisor Dr Afshin Parvardeh a parvardeh@sci ui ac ir 2010 MSC 60G99 62G10 62C10 Keywords Two sample hypothesis tests Nonparametric Bayes Dirichlet process Polya tree AbstractTwo sample comparison is an important problem in statistics such that in practical applications es pecially in biostatistics and laboratory data we are interested in doing it Namely given two sets ofsamples Y 1 F1 and Y 2 F2 where F1 F2 are unknown the purpose is to examine the equiv alent of F1 and F2 Classical nonparametric tests such as Kolmogorove Smirnov and Cram r vonMises do not scale well as the dimentionality increased General purpose in this thesis is to examinethe existence of systematic di erences between real valued responses under two di erent treatmentswithout considering any distribution for the data in Bayesian framework So Bayesian nonparametricapproaches would be used as a solution In this case since the parameter space is a high dimensionalparameter space and F1 and F2 are unknown so the examination of this problem is considered inthe Bayesian nonparametric area Such tests are encountered in various disciplines for example inmedical studies one may interest in the e ects of a new drug in K groups For more example seeBorgwardt and Ghahramani 55 Bayesian nonparametric is a fast developing area in statistics but there has been amount of workon Bayesian nonparametric hypothesis testing Bayesian parametric hypothesis testing where F1and F2 are known is developed in the Bayesian literature Bernardo and Smith 6 and most ofthe nonparametric work has been focused on testing a parametric model versus a nonparametricalternative Goodness of t test The rst work on Goodness of Fit problem Florens et al 22 Carotaand Parmigiani 13 used a Dirichlet process prior for the alternative distribution and compared to
استاد راهنما :
زهرا صابري
استاد مشاور :
افشين پرورده
استاد داور :
مجتبي گنجعلي، صفيه محمودي
