課程資料
國立清華大學
計算統計在資料科學中之應用
雷松亞 特聘教授
類別
A類
學分數
3 學分
開課人數
120
開放外校人數
10
授課語言
英文
上課日期
115.02.23~115.06.08
上課時間
週一09:00-12:00
上課地點
台積224
外校學生實體出席次數
16
外校學生實體出席日期
115.02.23~115.06.08
修課方式
校際選課
課程簡介
This course introduces computational approaches to statistical analysis and modeling for data science. Students will learn to write R programs that implement statistical concepts, simulate data, conduct inference and prediction, and produce reproducible, high-quality analyses for research and decision-making. Emphasis is placed on hands-on computation, algorithmic thinking, and professional data analytics coding standards.
課程內容
1. Computation and Statistics – Course overview, the role of computation in statistics, using R and RStudio.
2. Learning Computation – Exploration, inference, and prediction; social learning and peer review.
3. Description and Simulation – Descriptive statistics, kernel plots, simulating data and distributions.
4. Computational Intervals – Functions, iteration, resampling, and confidence intervals.
5. Computational Tests – Bootstrapping, classical hypothesis testing, simulation-based inference.
6. Nonparametric Testing – Bootstrapped hypothesis tests and empirical power.
7. Permutation Tests – Reshaping data, permutation methods, Wilcoxon test.
8. Multigroup Tests – ANOVA, Kruskal-Wallis, normality and familywise errors.
9. Inferring Relationships – Data similarity, correlation, and collaborative filtering.
10. Linear Regression Perspectives – Geometric and algebraic views of regression.
11. Applied Regression – Diagnosing nonlinearity and multicollinearity, stepwise modeling.
12. Moderation and Mediation – Indirect effects and bootstrapped tests.
13. Data Dimensions and PCA – Principal components, composites, and data reduction.
14. Structural Equation Modeling – Composite and factor models using SEMinR.
15. Machine Learning Methods – Prediction, cross-validation, decision trees, bagging, boosting.
16. Validation and Conclusions – Hyperparameter tuning, validation sets, and deployment tools.
2. Learning Computation – Exploration, inference, and prediction; social learning and peer review.
3. Description and Simulation – Descriptive statistics, kernel plots, simulating data and distributions.
4. Computational Intervals – Functions, iteration, resampling, and confidence intervals.
5. Computational Tests – Bootstrapping, classical hypothesis testing, simulation-based inference.
6. Nonparametric Testing – Bootstrapped hypothesis tests and empirical power.
7. Permutation Tests – Reshaping data, permutation methods, Wilcoxon test.
8. Multigroup Tests – ANOVA, Kruskal-Wallis, normality and familywise errors.
9. Inferring Relationships – Data similarity, correlation, and collaborative filtering.
10. Linear Regression Perspectives – Geometric and algebraic views of regression.
11. Applied Regression – Diagnosing nonlinearity and multicollinearity, stepwise modeling.
12. Moderation and Mediation – Indirect effects and bootstrapped tests.
13. Data Dimensions and PCA – Principal components, composites, and data reduction.
14. Structural Equation Modeling – Composite and factor models using SEMinR.
15. Machine Learning Methods – Prediction, cross-validation, decision trees, bagging, boosting.
16. Validation and Conclusions – Hyperparameter tuning, validation sets, and deployment tools.
授課教師簡介
國立清華大學服務科學研究所專任特聘教授雷松亞（Soumya Ray），畢業於威斯康辛大學麥迪遜分校，獲得博士學位。其研究領域為資訊系統，主要專注於使用者行為、具行動能力的人工智慧系統，以及計算方法。他教授軟體工程與計算統計等課程。目前擔任 [Information Systems Research] 與 [Journal of the Association of Information Systems] 期刊的編審委員，並擔任 [Decision Support Systems] 的副主編。
備註
更多的課程要求將於前三次上課時說明。
僅開放校際選課，不開放旁聽。
清華大學校際選課說明:https://curricul.site.nthu.edu.tw/
僅開放校際選課，不開放旁聽。
清華大學校際選課說明:https://curricul.site.nthu.edu.tw/