شماره ركورد كنفرانس :
3140
عنوان مقاله :
A Bayesian approach for modeling ceramic's fracture strength
عنوان به زبان ديگر :
A Bayesian approach for modeling ceramic's fracture strength
پديدآورندگان :
Yari Gholamhosein نويسنده School of Mathematics - Iran University of Science and Technology - Tehran - Iran , Saghafi Abolfazl نويسنده School of Mathematics - Iran University of Science and Technology - Tehran - Iran , Mirhabibi Alireza نويسنده Center of Excellence for Advanced Materials and Proces University of Science and Technology - Tehran - Iran , Ferizabadi Farahanio Ameneh نويسنده Tarbiat Modares University - Tehran - Iran
كليدواژه :
Weibull distribution , Fracture strength , Monte Carlo Method , Bayes Estimation sampling-importance-resampling
عنوان كنفرانس :
يازدهمين كنفرانس آمار ايران
چكيده لاتين :
In modeling fracture strengths, a researcher may have rough predictions about the characteristics of his trial. Besides, some prior knowledge could also be available from historical data. In this paper, this prior knowlEdge is used via Bayes rule to build more reliable Estimations of parameters of the Weibull distribution which is widely employed in reliability and lifetime studies. This approach is really helpful especially when dealing with Small sample sizes. After applying different priors, the sampling-importanceresampling technique, which is a special Markov Chain Monte Carlo method. is used to study the posterior properties. Some num simulations are given to Compare the statistical properties of Estimations by the Bayes and the common maximum likelihood and linear regression methods.
شماره مدرك كنفرانس :
4219389
