PhD Course: Behavioral Economics, Learning, and the Big Data Revolution
Speakers: Professor Ido Erev and Dr. Ori Plosky, Technion, Haifa, Israel and Associate Professor Davide Marchiori, SDU
Hosted by: SDU, Department of Business and Economics
Dates: September 23-26, 2019
Locations: SDU Campusvej 55, Odense M
Rooms: Sept. : 23 O96 & O94, Sept.: 24 O95, Sept.: 25 O95, Sept.: 26 O97
Registration: Behavioral Economics, Learning, and the Big Data Revolution
Background: The course focuses on human reaction to economic incentives, and the way the Big Data revolution changes the study of this topic and its implications. It considers the two-sided relationship between Behavioral Economics and the Big Data revolution. The first focuses on how our understanding of human reactions to incentives can help facilitating the positive effects of this revolution and minimizing its negative effects. The second focuses on the different ways new Big Data techniques can improve our ability to predict the effects of incentives on human behavior.
Course topics and procedures: Each of the four days will start with a review of the basic research on the topic and concludes with hands-on practice toward a team project.The students will be able to choose between two type of projects: Projects that involve mastering the use of computer simulations and machine learning tools (for students with basic knowledge of R or Python), and projects that focus on the design of effective incentive systems. The division to sessions is described in the following table.
Topics and program for the day:
|
Day |
Topics |
Recommended reading |
|
Sept. 23 |
Rational choice and the effect of experience Lecture: Review of research of the conditions under which experience increases maximization of expected return, and the implications of this research to the design of effective big data technology.
Optional practice: The use of computer simulations to develop models that approximate human reaction to incentives. Examples include: FP, basic Reinforcement learning, EWA, naïve sampler.
|
|
|
Sept. 24 |
Human and Machine learning Lecture: Introduction to machine learning, and the relationship between human and machine learning.
Optional practice: The use of basic machine learning tools to predict economic behavior. The development of predictors for the 2020 “Choice Prediction Competition” Students will be able to develop their model using simulation, machine learning, or both methods. |
|
|
Sept. 25 |
From anomalies to predictions Lecture: The classical deviations from rational choice and the impact of experience. Optional practice: Deriving the prediction of descriptive model of decisions under risk including prospect theory. Features engineering. |
|
|
Sept. 26 |
Practical implications and student presentations Practice: Students’ presentation |
|
Learning Objectives
The course will introduce students to the main findings in experimental studies of human reaction to economic incentives.
In addition, the course will provide opportunities to master the use of computer simulations and machine learning tools, to develop models, derive quantitative predictions, and address practical prediction tasks.
Instructors:
Main instructors: Professor Ido Erev and Dr. Ori Plonsky (Technion, Israel), Associate Professor Davide Marchiori (University of Southern Denmark)
Admission procedure: Students who would like to participate in the course should send a short letter of motivation (max 1 page) describing (i) their research interests and (ii) why they are interested in participating in the course. Letters should be sent to sidg@sam.sdu.dk (Subject: “PhD course on Behavioral Economics, Learning, and the Big Data Revolution”) no later than August 28, 2019.
The slots available are limited and acceptance will be communicated asap after the deadline.
Prerequisites: basic knowledge of Microeconomics and Game Theory. Knowledge of R or Python is a bonus as some (optional) activities will require programming.
ECTS points and course evaluation: The evaluation will be determined by a final project. The students will choose between a project on computer simulation (participation in a choice prediction competition) or a more traditional final paper.
Upon completing all course activities (attendance of the entire course + participation in course activities + submission of the project), participants will be awarded 4 ECTS credits and a course certificate.
Course fees:The course is free of charge for PhD students in Economics from Scandinavian Universities (Denmark, Sweden, Norway, Finland, Iceland). The fee for other students is of 100 Euros per day. Students are expected to cover their expenses. Lunches + a course dinner will be included.
Course Website: Click here
Course responsible: Associate Professor Sibilla Di Guida, Department of Business and Economics, University of Southern Denmark
Reading
- Breiman, L. (2001). Statistical modeling: The two cultures (with comments and a rejoinder by the author). Statistical science, 16(3), 199-231
- Erev, I., Ert, E. Plosky, O., Cohen, D., & Cohen O. (2017). From anomalies to forecasts: Toward a descriptive model of decisions under risk, under ambiguity, and from experience. Psychological Review, 122(4), 621.
- Erev, I., & Roth, A. E. (2014). Maximization, learning, and economic behavior. Proceedings of the National Academy of Sciences, 111(Supplement 3), 10818-10825.
- Kahneman, D., & Tversky, A. (1979). Prospect theory: An analysis of decision under risk. Econometrica,47, 263-291.
- Pinker, S. (2011). Decline of violence: Taming the devil within us. Nature, 478(7369), 309.
Plonsky, O., Teodorescu, K., & Erev, I. (2015). Reliance on Small Samples, the Wavy Recency Effect, and Similarity-based Learning. Psychological Review, 122(4), 621–647.