贝叶斯计算统计

贝叶斯计算统计（48课时/3学分）

Computational Bayesian Statistics (48 hours/3 credits)

课程描述：

本课程旨在培养学生掌握贝叶斯推断的理论知识与实践技能，使其能够广泛应用于实际场景。课程首先介绍贝叶斯框架的基本原理，随后重点解析基于马尔可夫链蒙特卡洛（MCMC）的核心推断算法，并结合实际案例探讨多种经典模型。具体内容包括：‌贝叶斯统计学导论；‌模型检验、比较与选择；贝叶斯计算入门；‌吉布斯采样；‌Metropolis-Hastings算法‌；‌缺失数据技术；‌分层模型‌与‌回归模型‌；‌高斯过程模型‌。

‌ 

The aim of this course is to equip students with the theoretical knowledge and practical skills to perform Bayesian inference in a wide range of practical applications. Following an introduction to the Bayesian framework, the course will focus on the main Markov chain Monte Carlo algorithms for performing inference and will consider a number of models widely used in practice. Topics covered are: Introduction to Bayesian statistics; model checking, comparison and choice; introduction to Bayesian computation; Gibbs sampler; Metropolis-Hastings algorithm; missing data techniques; hierarchical models; regression models; Gaussian process models.