課程名稱 |
應用選擇資料分析 Applied Discrete Choice Analysis |
開課學期 |
110-2 |
授課對象 |
生物資源暨農學院 農業經濟學研究所 |
授課教師 |
石曜合 |
課號 |
AGEC7144 |
課程識別碼 |
627EM5290 |
班次 |
|
學分 |
3.0 |
全/半年 |
半年 |
必/選修 |
選修 |
上課時間 |
星期四3,4,5(10:20~13:10) |
上課地點 |
共207 |
備註 |
本課程以英語授課。建議具備計量經濟相關知識, 限碩士班以上 總人數上限:12人 |
課程網頁 |
https://cool.ntu.edu.tw/courses/4471/assignments/syllabus |
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 |
Discrete choice models (DCM) have been widely applied to study individual choice behavior problems in many fields such as economics, marketing, environmental management, and transportation. This course will mainly focus on the methods and applications of DCM on topics in agricultural and environmental economics. For example, how people choose their vacation destinations, select the ideal environmental programs/policies, and pick the food product they want. |
課程目標 |
The course first introduces the theories and framework of DCM, data collection for DCM, as well as how DCM has been applied in various disciplines. The course covers some of the fundamental discrete choice models (logit, nested logit, probit, and mixed logit) and includes lab sessions where students can learn how to analyze how people making choices using real datasets. The principle software used is R. The primary goal of the course is for students to gain hands-on experience in using discrete choice techniques for practical applications.
This course is not simply about how to analyze discrete choice data nor how to use R, but about the entire process of conducting empirical projects using discrete choice models. The objectives of the course are helping students to develop the ability to explore questions, find data, analyze the data with appropriate methods, interpret the results, and justify the contributions. |
課程要求 |
* In-class participation: 30%
* Paper review and seminar: 30%
* Project proposal or Replication project: 40%
Students are expected to attend regular class time lectures and lab sessions. In addition, students are expected to actively participate in class and complete a paper review and a project proposal. Problem sets will be assigned throughout the semester for practice.
Students are expected to write a brief project proposal for an empirical choice analysis project. The proposal has to address the following: (1) what the research question is, (2) what have been done in the literature, (3) how the data can be collected (or where the data are), (4) the analytic framework and models, and (5) the expected findings and why the question and findings are interesting and/or useful. Students are strongly encouraged to meet with the instructor before the mid-term to discuss the proposal's topic. The proposal is due at the end of 6/2 (the final week). The instructor will email the feedback for the proposal within 2 weeks after the due date.
Project proposal may be substituted by replicating the results of a paper listed on the reading list. The grade will be largely based on how well the results are replicated. In the most ideal cases, students are expected to reproduce ALL tables in the main text. The numbers should match the original results to a certain level of precision. That being said, students are not likely to get a good grade by producing a table of summary statistics and another table with only a mixed logit model. In case some of the results cannot be replicated, especially because the models are not included in any R packages (so one needs to write the MLE functions or such), students would want to use the best available models and discuss the causes of the discrepancies. If students want to replicate the results of a paper that is not on the list, please email the instructor. The replication project is also due at the end of 6/2 (the final week).
The work outside of class is typically three to six hours per week. Grading for the class will follow the University's ranking and percentile score system.
The instructor encourages feedback throughout the semester to make sure the course goals and students' expectations are being met. |
預期每週課後學習時數 |
|
Office Hours |
|
指定閱讀 |
The following books are recommended for reference.
* For discrete choice modeling
Champ. P. A., Boyle., K. J., & Brown T. C. (Eds.) (2017). A Primer on Nonmarket Valuation, 2nd Edition. Dordrecht: Springer Netherlands. (Available via NTU library eBooks)
Mariel, P., Hoyos, D., Meyerhoff, J., Czajkowski, M., Dekker, T., Glenk, K., ... & Thiene, M. (2021). Environmental valuation with discrete choice experiments: Guidance on design, implementation and data analysis. Springer Nature. (Open access)
Kleiber, C., & Zeileis, A. (2008). Applied Econometrics with R. Springer Science & Business Media. (Available via NTU library eBooks)
* For R programming and other relevant topics (e.g., machine learning)
R Cookbook: https://rc2e.com/
Hands-On Programming with R: https://rstudio-education.github.io/hopr/
R for Data Science: https://r4ds.had.co.nz/
Hands-On Machine Learning with R: https://bradleyboehmke.github.io/HOML/
An Introduction to Statistical Learning: http://faculty.marshall.usc.edu/gareth-james/ISL/
* Course handouts, slides, and data will be made available at the course website by the instructor. |
參考書目 |
Train, K. E. (2009). Discrete Choice Methods with Simulation, 2nd Edition. Cambridge University Press. (Available online at https://eml.berkeley.edu/books/choice2.html ) |
評量方式 (僅供參考) |
|
週次 |
日期 |
單元主題 |
Week 1 |
02/17 |
Course overview: the science of choosing |
Week 2 |
02/24 |
Properties of discrete choice models; Random utility model |
Week 3 |
03/03 |
The fundamental in discrete choice: Logit |
Week 4 |
03/10 |
When IIA fails: Nested Logit and GEV models |
Week 5 |
03/17 |
Accommodating random preference heterogeneity and repeated choices: Probit |
Week 6 |
03/24 |
The current standard in DCM: mixed logit (I) |
Week 7 |
03/31 |
Mixed logit (II) and count models |
Week 8 |
04/07 |
Mid-term (No class) |
Week 9 |
04/14 |
DCM in ag and env econ (I): Revealed preference
* Seminar paper selection due |
Week 10 |
04/21 |
DCM in ag and env econ (II): Stated Preference
* Final proposal topic due |
Week 11 |
04/28 |
Student seminar(s)
Nudging Energy Efficiency Behavior: The Role of Information Labels (陳禹嫺)
Risk Preferences, Risk Perceptions, and Flood Insurance (傅羿寧) |
Week 12 |
05/05 |
Ordered choices and outcomes
Some alternatives to RUM
Student seminar(s)
Sad or Happy? The effects of emotions on stated preferences for environmental goods (王冠傑)
An experiment on the vote-buy gap with application to cage-free eggs (吳子欣) |
Week 13 |
05/12 |
Endogeneity in DCM and other issues in DCM
Student seminar(s)
Using revealed preferences to estimate the value of travel time to recreation sites (蕭雲豪) |
Week 14 |
05/19 |
Statistical learning and discrete choice analysis
Proposal Presentations |
Week 15 |
05/26 |
Proposal Presentations |
Week 16 |
|
Final Week (no class) |
|