Eco326: Information Economics

Syllabus

Homework 1 and Solutions.
Correction: The solution to 3(b) should say that there is trade if r<600, not r>550. This means that 60% of seller trade.
Clarification: Several people have asked about the difference between common and private values. With common values (e.g. Q2) the seller's private information (&theta) affects the buyer's valuation. With private values (e.g. Q3) the buyer knows her valuation and it is not affected by the seller's private information (r).
Clarification: In solution to 1 I assumed that the captains care about the expected 'p' of the surgeon. However all that is required is that both captains have the same risk profile.

Midterm 1 and Solutions. (Mean 33, Max 50). Percentage grade approximately equals 1.5×(grade+8).

Homework 2 and Solutions.
The first three questions are of test difficulty. The next three are trickier but are a good test of whether one understands what's going on.

Midterm 2 and Solutions. (Mean 27, Max 46). Percentage grade approximately equals 1.5×(grade+17).

Homework 3 and Solutions.
Typo: In question 6(b) the relation should be &theta_t+1= p_t+1.
Observation: In question 4(c), this is not literally the pivitol mechanism, as in the previous questions. Rather, it in in the style of the pivitol mechanism.

Final office hours: Thursday 14th April, 2-4pm (NOTE CHANGE). If you cannot attend please let me know: I am happy to extend these hours. These office hours will go on until you have run out of questions. This will also be your last chance: I am unlikely to be able to see anyone after this date.

Final and Solutions.
Tuesday 19th April, 9-11am. Room NE160. This exam is open book/notes.

For the final, the following is a rough ranking of importance

1. akerlof, spence and monopolistic screening.
2. clarke-groves mechanisms, competitive screening, moral hazard.
3. credit rationaing, cheaptalk, first-price auctions,

Some students find it tricky to adjust to the style of this class from eco200. Here are some pointers which might be useful, although different people have different styles.

Understanding the material

• Don't memorise. The aim is to understand the model - you cannot possibly memorise everything I can ask you.
• Play with the model. You should try to become best of friends. As with all your friends, try smacking it around. What makes the model work? What are the crucial assumptions? When will it not work?
• One key step is to go from intuitively knowing why something is true or false to being able to construct a logical argument. Try to run the explanation past a friend, perhaps in the context of a study group. Ask your friend to be skeptical: it they don't understand a step then you need to be able to explain it more clearly.
• One of the biggest problems is knowing when you really know something, and when you don't. It's easy to convince yourself you understanding is better than it is (particularly if you have a good memory). Be honest with yourself.
• Read critically. Don't worry if a page in a book takes four hours to read. When you don't understand a sentance stop and think about it. Always have a pad of paper next to you when reading and try and check that every line in an argument is correct.

In the exam

• Don't Panic! Try to break the problem down into its component parts.
• If you can't think how to write the answer mathematically, write it in English. In fact, providing intuition is always a good idea - if something goes wrong in your answer, I'll know whether you understand what's going on.
• Answer the question. There are two separate but related problems. First, students often miss parts of questions - yet this is just throwing marks away. Please read the question carefully (and several times) to make sure you are not missing part of it. My advice is to number each part (i.e. anything with a question mark) when reading the question and then check them off when you are done. Second, students often answer what they want the question to be (e.g. one on the last homework), not what the question actually asks. Please do not fall into this trap - I don't given marks for wishful thinking. It is better to give a half decent answer to the actual question than a perfect answer to a fictional question.

Index

Date: 30 March, 2005