Sotiris Georganas - Σωτηρης Γεωργανας - Associate Professor - Economics - Auctions, Experiments, Bounded Rationality

Lecture 5 - Collusion
Problem Set 5 without Solutions / Problem Set 5 with Solutions
If you have nothing better to do, have a look at the movie "The informant!". It is funny and actually based on the real story of firms colluding in the 1990s to fix the price of lysine (an additive used extensively for animal feed).

Midterm exam - Solutions
If you want some more exercises to work on, try this:
Problem Set 5b with Solutions -->

Lecture 6 - Advertising
Problem Set 6 and Bonus Signalling game: without Solutions / Problem Set 6 with Solutions
Small Notes on Signalling Games
For more on signalling games check Lecture 8 from the Game Theory class.

Lecture 7 - R&D
Problem Set 7: without Solutions - Solutions

ONLINE assignment with Solutions (warning: the sequence of questions might be different for some people to the one they actually solved, but the material is the same)

Lecture 8 - Behavioural IO (ppt)

Industrial Economics - MSc - EC5020

Lecture 5 with solutions

Lecture 6 - Mergers

Lecture 7 - Repeated Games
Problem Set with Solutions

Lecture 8 - Product Differentiation

Lecture 9 - Experimental Markets

Experimental Economics

Questionnaires - Experiments

Overconfidence Questionnaire

Anchoring & Law of Small Numbers Questionnaire

Game Theory Questionnaire

Covid Questionnaire

Games and Experiments (Veconlab)
Choose login as participant
>Initial login
>Wait for me to tell you the session ID
>Enter your data and a simple password you will remember, to login again in case the experiment crashes
>Read the instructions and play!

Lecture material

⇒ Lecture 1

⇒ Lecture 2
papers to read:
-An Experimental Study of Sequential Bargaining (1989) Jack Ochs and Alvin E. , AER
-In Search of Homo Economicus: Experiments in 15 Small-Scale Societies (2001) Henrich, Boyd, Bowles, Camerer, Gintis, McElreath & Fehr

Fun stuff: Monkeys reject unequal offers - apparently monkeys are not very good economists either

⇒ Lecture 3 - Public Goods & Cooperation
papers to read:
-Fischbacher, Urs, and Simon Gachter (2010) "Social Preferences, Beliefs, and the Dynamics of Free Riding in Public Goods Experiments." American Economic Review, 100(1): 541–56.
-Urs Fischbacher, Simon Gächter & Ernst Fehr (2001) Are people conditionally cooperative? Evidence from a public goods experiment, Economics Letters, Volume 71, Issue 3, June
Matlab Code for Conditional Cooperator Simulations

⇒ Lecture 4 - Ind. Decision Making

⇒ Lecture 5 - Experimental Markets - Double Auctions

⇒ Lecture 6 - Bubbles
papers to read:
-Nonspeculative Bubbles in Experimental Asset Markets: Lack of Common Knowledge of Rationality vs. Actual Irrationality (2001) Vivian Lei, Charles N. Noussair and Charles R. Plott, Econometrica
-Traders’ Expectations in Asset Markets: Experimental Evidence (2007) Haruvy, Ernan, Yaron Lahav, and Charles N. Noussair, AER

⇒ Lecture 7 - Single Object Auctions
papers to read:
-The Winner's Curse and Public Information in Common Value Auctions (1986) John H. Kagel and Dan Levin, AER
-Last-minute bidding and the rules for ending second-price auctions: Evidence from eBay and Amazon auctions on the Internet (2002) Alvin E. Roth and Axel Ockenfels, AER

⇒ Lecture 8 - Noisy behavior, quantal response equilibria and level-k
papers to read:
-An Experimental Study of the Centipede Game (1992) Richard D. McKelvey and Thomas R. Palfrey, Econometrica
-English Auctions with Resale: An Experimental Study (2011) GEB
Matlab code to find QRE in symmetric normal form games. You can also use Gambit for normal form and extensive form games, of a reasonable size.

Instructions for your experimental project

Useful material: A simple 20 minute presentation of an experimental paper.

Here are the 3 main things that I will consider when reading the project (in parentheses the approximate weight).
-Novelty (35%)
-Effort (30%)
-General economic importance of the issue (35%)
Prerequisite: showing that you know the literature (related to effort). Doing something that somebody else did, without mentioning it, means you were either lazy or cheating. That's bad. If you do something that somebody else did and do mention it, it just makes for a boring paper. Try to avoid it.

How to search through the literature:
To check whether one of your ideas exists in the literature it is a good idea to use Google Scholar. Search for basic keywords, e.g. "risk aversion experiments". Scholar will show you a bunch of papers, ranked by how important it thinks they are. Identify a couple of major papers from this list. If what you want to do does not appear in these papers, check whether newer papers have actually covered the topic. A way to search for them is to look for papers that are citing this major paper by clicking on the link below the abstract. You can restrict the search to newer papers, to make sure you don't miss the latest important contributions.

Length:
The length of the final paper is not important, like I told you before, Nash's papers were small but extremely important. As a guide, it's probably good to be in the 5-15 pages range. The structure of the paper should broadly follow the structure of one of the papers you have read during the class, obviously adapted to the type of paper you are writing (actual novel experiment? simulation? mostly a critique of older papers?)

Deadlines: the deadline for the submission of an outline is the 15th of January. The deadline for the actual paper is the 25th of May (but don't wait until the last minute!). Let me know if you have problems.

Dissertations

Structure

Even for literature reviews, I would like to see a summary of theoretical models first, and then summarise the empirical findings.

Literature

Here are some tips how to search through the literature and find interesting papers.

Start by searching in the database of your choice (Google Scholar, repec, jstor...). Use basic keywords, don't make it at first too specific. Now, when you start finding papers, which ones to read and use?

First and foremost:
Not all papers are created equal.

Try to use good papers in good journals

What is a good journal? People in the profession sort of know what the good journals are, through experience. Since you do not have much experience, you can use ranking lists. Example: the ABS list

Usually, a paper in a 3 or 4 star (“grade 3” or “grade 4”) journal is good, lower tiers will tend to publish more boring papers.

Now, suppose you have found a paper published in such a journal. Is it good? Maybe. Citations are a good proxy for the quality of a paper, especially if it has been published more than 5 years ago. The easiest way to check for citations is through Google Scholar.

Using old papers (say 20 years old) is not a terrible idea, if they are influential (i.e. they have maaaany citations). A note about citations: some fields tend to get more citations than others, empirical work maybe more than theoretical. When you are looking at citations, make sure you compare to similar papers in similar fields.

Now, once you have found a paper, always check for newer takes on the main ideas of the paper, maybe major points of the paper are obsolete by now. An easy way to do this is by checking papers that are citing this particular paper (just click on the “Cited by XX” link in Scholar).

Read papers properly

And I mean this. Using a paper you have not understood at all is a bad idea.

Summarise properly

Do not just copy the abstract, or select random statements from the introduction. You have to summarise the main points of the paper, as they relate to your work and other papers you are citing.

Cite them properly when summarizing.

Make sure that you name the papers you are using, and if using their own words make sure you use quotes.