14.12  Econ Apps of Game Theory

Final announcement ~!

Dear all,

If you have not done so, please take a minute to fill out the course evaluations. The TAs would very much appreciate your feedback. Thanks!!

Unless something unexpected happens, this is the final announcement of the course :-). Good luck on the final.

Best regards,
Kevin.

Announced on 18 December 2017  12:32  a.m. by Kevin Kainan Li

Piazza questions & Final Exam

Dear all,

I'm going through Piazza now and going through all the unanswered questions (and requests for solutions) which have been piling up.

If you have a lingering question, now's the time to ask! That is, if you post on Piazza/email me before (roughly) midnight , I'll make sure to answer within the hour/as time allows.

As a reminder, the final exam time & location is

Monday (tomorrow) 9:00AM - 12:00PM @ Walker Gym.

Best regards,
Kevin.

Announced on 17 December 2017  9:56  p.m. by Kevin Kainan Li

Pre-Final Grades

Dear all,

As I mentioned during the review session, I just uploaded a 'dummy' assignment called "Pre-Final Grades". This is the weighted average of your grades so far. See below for details.

If this number is not what you expect, please email me ASAP.

The "Pre-Final Grade" X is calculated as follows:

X = 1/3 M1 + 1/3 M2 + 1/3 H,

where M1 and M2 are your midterm 1 and midterm 2 grades (the 15% bonus for the Conflict midterm 1 is included), and H is the average of your highest 4 homework grades,

H = 1/4 (HW1 + ... + HW5 - min(HW1,..,HW5)). (The potential bonus of Homework 2 is included.)

The (1/3, 1/3, 1/3) weight is due to midterms and (cumulative score) of homeworks having equal weight of 20%. Therefore, if your "Pre-Final Grade" is X and your final exam score is Y (out of 100), your final numerical grade (out of 100) of the class is given by:

0.6 X + 0.4 Y.

Don't hesitate to email me if you have questions.

Best regards,
Kevin.

Announced on 17 December 2017  9:28  p.m. by Kevin Kainan Li

Pset 5 graded

Hi all,
I have graded the final Pset. They are in the relevant place in E52-300. Sorry for the delay. As in other problem sets, grading has been very lenient. However, please make sure you know the correct answer. The answer for a large majority is not perfect for questions 4 and 5. I will like to mention about the major mistakes since most probably won’t be able to collect the graded scripts now:
1) Q1 & Q2: Remember BNE involves functions from Types to set of actions. Also, note that in the second problem Expectation of a function of theta2 given theta1 is a function of theta1.
Also, while defining a Bayesian game, you should define the Joint distribution on the type space. Mere marginals are not enough (if independent, say that they are independent, and hence, we know the Joint distribution also)

2) Q3: Buyer’s strategy is defined as a function of beta and not vB. The buyer doesn’t know the latter. Also, while calculating the integrals, you should not in any of these steps substitute E[vs]=0.5, etc. The good won’t be sold for any value of vs.
3) Q4: Most of you have not mentioned about beliefs of Sally. Also, even if you have mentioned about beliefs for p in [0, (1/(2-alpha)], for completeness, it is also good to mention about beliefs for p>(1/(2-alpha), i.e. for off the equilibrium.
4) Q5: Please see Kevin’s comments in the solution for a perfect answer. It is very important to mention about Bob’s strategies and beliefs for prices off the equilibrium. You should also verify Consistency.
I will try to send you a scanned copy of your script if you request me in my e-mail (souravs@mit.edu)

Announced on 17 December 2017  4:29  p.m. by Sourav Sarkar

Time + Location for Final Exam Review Session

Dear all,

The voting on Piazza and the Doodle poll is now closed. It seems that the original 4-6pm time slot won by one vote (!).

The final exam review session will be held at this time and place:

Date & Time: Friday (12/15), 4-6PM.
Location: E51-395, (same place as the afternoon recitations, also the same place as midterm 2 review).

We'll be going over examples of finding Bayesian Nash equilibria and sequential equilibria, with problems drawn from past finals and lecture notes. Post on Piazza or email me if you would to see other stuff covered.

Since Wednesday is the last day of class, there won't be office hours on Thursday. I'll stick around after the review session on Friday if you need to talk to me in person. Otherwise, post on Piazza or send me an email.

Best regards,
Kevin.

Announced on 13 December 2017  2:32  a.m. by Kevin Kainan Li