home           cv           research           teaching

Behavioral and Experimental Economics
Fall 2020, Higher School of Economics

The class will meet online this semester! Please e-mail me if you are registered and need the link.

Syllabus.

September 9: Introduction

Class materials: Slides. Hertwig and Ortmann (2001) (optional).
Homework: None.

September 11: Our first experiment

Class: Introduction (continued). Our first experiment.
(2) Read Oprea and Yuksel (2020). Focus on the introduction up to the literature review (up to ''our paper contributes to several literatures'') and the experimental design (pages 7-12).
Written assignment Question 1 (one paragraph): Come up with a hypothetical real life example of how social exchange makes motivated beliefs more biased. Make sure to explain what the belief is about, how and where it is being exchanged, and why you think it will become more biased as a result of the exchange. Be specific and don?t take an example already discussed in the paper!
Written assignment Question 1 (one paragraph): Oprea and Yuksel define motivated belief as believing in something because you want it to be true. Does the capital punishment experiment described in the article involve motivated beliefs? Why or why not? Explain!
Important: Submit your assignment by 13:00 on Friday, September 18 to pevdokim@gmail.com with the subject line EE2020 HW1.''

September 15: Bayesian Learning

Class materials: Slides.
Homework: None.

September 18: Bayesian Learning (continued)

Class materials: Slides. How to read a medical test: an article that come out this week in the New Yorker that discusses many of the concepts we talked about in class, including the sensitivity/specificity of a test, base rate, the positive predictive value of a test (not just in the context of coronavirus), and the importance of Bayesian reasoning for doctors.
Homework: The homework assignment is to run your first experiment. The experiment I want you to run is a learning experiment similar to the two urn experiment your participated in in class, but conducted using pencil and paper. The experimental setup is as follows:
You (the experimenter) and the participant start by sitting across each other from the table. You read the instructions to the participant. The instructions can be in Russian or English, your choice. After reading the instructions, the experiment begins.
The experiment consists of a game played 5 times. In the game are two groups of cards. Group 1 consists of 2 "A" cards and one "B" card. Group 2 consists of 2 "B" cards and 1 "A" card. At the beginning of each game, you flip a coin, which determines which group of cards you use. You ask the participant to turn away during the selection phase.
After the group of cards is selected, you put the selected group of cards into a hat. In each period of the game, you a draw a card from the hat (with replacement) and ask the participant to report what he/she thinks the chances in 100 are that the selected group consists of two A cards and one B card. At the end of each period, you write down in this document what the selected group was in the game, what card was drawn, and what belief the participant reported. At the end of the fifth period of the game, after the participant reports his or her belief, you reveal what group of cards was selected. You then flip a coin again to determine the group of cards and continue to the next game.
Assignment, part 1: Write down your own instructions for this experiment. These are the instructions I want you to read to the participant and send to me as part of the assignment.
Assignment, part 2: Conduct the experiment. Record the experiment on video and submit the video to me as part of the assignment. You can send the video directly if the file is small enough or use an online storage service such as Google Cloud or Dropbox.
Assignment, part 3 Submit the data for the experiment in a CSV document. The document should be in this format (i.e., fill out the cells in the blank csv document in the link).
Assignment, part 4 Analyze the data. Specifically, I want you to estimate the following model:

## $$log \bigg( \frac{p_t}{1-p_t} \bigg)= \alpha \cdot log \bigg[ \frac{ P(s_t| A\;group) }{ P(s_t| B\;group )}\bigg]+ \beta \cdot log \bigg[ \frac{p_{t-1}}{1-p_{t-1}} \bigg] + \epsilon_t$$

Report the coefficients and interpret them. Do you find evidence of base rate neglect in your data?
Important: Submit your assignment by 13:00 on Friday, September 25 to pevdokim@gmail.com with the subject line EE2020 HW2.'' Your submission should consist of 4 files: the instructions file, the video file with the recording of the experiment, the CSV file with the data, and a document summarizing your estimation results/findings on base rate neglect.

September 23: Bayesian Learning: Working with Data

Class materials: Slides. Files for the COVID survey (data, DO file).
Homework: See September 18.