課程名稱 |
貝氏空間分析 Bayesian Analysis for Spatial Data |
開課學期 |
112-1 |
授課對象 |
共同教育中心 統計碩士學位學程 |
授課教師 |
溫在弘 |
課號 |
Geog5140 |
課程識別碼 |
228 U3530 |
班次 |
|
學分 |
3.0 |
全/半年 |
半年 |
必/選修 |
選修 |
上課時間 |
星期三7,8,9(14:20~17:20) |
上課地點 |
地理二教室 |
備註 |
本課程中文授課,使用英文教科書。工程與環境統計領域選修課程之一。 總人數上限:20人 |
|
|
課程簡介影片 |
|
核心能力關聯 |
本課程尚未建立核心能力關連 |
課程大綱
|
為確保您我的權利,請尊重智慧財產權及不得非法影印
|
課程概述 |
This course aims to provide an introductory exploration of Bayesian spatial statistics, with the goal of equipping students with the necessary computational skills to analyze geographically represented data. Bayesian inference has become one of widely utilized statistical methods for addressing spatial and temporal dependency in the fields of geo-informatics, environmental science, and social sciences. Throughout the course, we will focus on Bayesian modeling techniques that are specifically designed for spatial-temporal data in geography and related scientific disciplines. The lectures will be organized around three main themes: the fundamental concepts of Bayesian inference, generalized hierarchical models, and conditional autoregressive models. It is expected that students have a prior background in statistics, including regression analysis as well as a basic understanding of statistical computing. The course will involve substantial computational work, primarily utilizing the R programming language. Additionally, students are also encouraged to possess a background in geographic information systems (GIS). |
課程目標 |
Students should be able to: [1] understand the concepts of Bayesian inference, [2] understand how spatial autocorrelation plays a role in statistical modeling, and [3] use Bayesian statistical models for their own research and implement them using the R language. |
課程要求 |
Course Participation, Computer Labs and Weekly Assignment |
預期每週課後學習時數 |
|
Office Hours |
|
指定閱讀 |
Robert P. Haining, Guangquan Li (2020), Modelling Spatial and Spatial-Temporal Data: A Bayesian Approach, Chapman and Hall/CRC |
參考書目 |
|
評量方式 (僅供參考) |
No. |
項目 |
百分比 |
說明 |
1. |
Mid-term Exam |
25% |
|
2. |
Final Exam |
25% |
|
3. |
Labs and assignments |
50% |
|
|
針對學生困難提供學生調整方式 |
上課形式 |
提供學生彈性出席課程方式 |
作業繳交方式 |
學生與授課老師協議改以其他形式呈現 |
考試形式 |
|
其他 |
由師生雙方議定 |
|
週次 |
日期 |
單元主題 |
第1週 |
09/06 |
Course Introduction |
第2週 |
09/13 |
Sampling: Monte Carlo Methods |
第3週 |
09/20 |
Bayesian Inference: Binomial Probability Model |
第4週 |
09/27 |
Bayesian Conjugate Priors |
第5週 |
10/04 |
Markov Chain Monte Carlo (MCMC) |
第6週 |
10/11 |
MCMC Sampling and Diagnostics |
第7週 |
10/18 |
Using RStan for Posterior Inference and Prediction |
第8週 |
10/25 |
* Midterm Exam * |
第9週 |
11/01 |
Bayesian Linear Regression |
第10週 |
11/08 |
Generalized Linear Model |
第11週 |
11/15 |
Hierarchical Linear Models: Estimating Random Effects |
第12週 |
11/22 |
Spatial Neighbors and Autocorrelation |
第13週 |
11/29 |
Conditional Autoregressive Models |
第14週 |
12/06 |
Estimating Localized Spatial Autocorrelation |
第15週 |
12/13 |
Spatio-temporal Modeling |
第16週 |
12/20 |
* Final Report * |
|