Topic outline

  • Thinking through and working on problem sets is the most important learning method in this course. Most of the “self-study” in the estimated workload (see course syllabus) consists of working on problem sets. The schedule of problem set deadlines is available in the course schedule, which is linked from the course home page. (Working on problem sets is also the best single method of preparing for the exam.) While there is much variability between problems, on average one numbered problem in a set is expected to take about 1-2 hours to complete.

    Homework solutions to each problem set must be submitted in one pdf-file (unless otherwise noted). Add your clearly typed student ID on the top of the first page, this makes graders’ work easier. Make sure to test your scanning quality well before the first deadline. 

    You are allowed and indeed encouraged to discuss the problem sets with fellow students taking this course. However, each solution must be individually written up (with the exception of specifically pointed out Group work). 

    For a solution to be acceptable it must include the explanation behind your reasoning. A mere bottom line answer is not acceptable. You don't need to outline all intermediate steps in a calculation (even if the model solution does) but you do have to explain the rationale behind your calculations. Why did you set up this computation, and what did you find out with it? You can use calculators and computers to aid you as much as you want and also to produce graphs. 

    When asked to give your own example then do not use examples that are very close to those used in class or in the textbook.

    Weekly schedule

    Problem set questions are based on material from previous weeks' course topics, with the exception of the last problem set which is due on the day after the last lecture. 

    There is a 5 point bonus for submitting your assignment in time for the recommended deadline, which is usually on Wednesday 15:00. You can update your submission until the final deadline, but notice that the time of your final revision is your official submission time. The bonus cannot increase total points above 100, but will "make up" for any lost points in the same problem set.

    The final deadline is on Thursday exactly 24 hours after the recommended deadline. This is also the time when model solutions are published at MyCourses. After the final deadline submission becomes impossible -- later submissions will not be accepted. 

    Points earned from each question will be available usually within one week. Questions regarding the grading should be addressed to the teaching assistant. 

    A small number of interactive problems (at most 15% of total) will be done in class. They are graded separately, with the answers typically submitted during class.

    Standard interpretations

    The features explained in the initial part of a problem before any possible subparts (a, b, c…) apply to the entire problem unless otherwise noted. The features in each subpart only apply to that subpart, unless otherwise noted. If there are further instructions after the list of all subparts these also apply to the entire problem unless otherwise noted. Maximum points for each question are shown in the problem set, they are equally divided between the subparts, unless otherwise noted.

    Unless otherwise stated, you can assume that decision-makers are risk neutral and maximize the present value of their own payoffs (typically profits for firms, utility or consumer surplus for consumers). For discounting purposes, you can assume that future periodic payoffs are realized at the end of a period. In particular, period t=0 refers to now immediately ("current period", "this year", etc), period t=1 to the end of first period, that is, "one period from now", "next year", etc. 

    Even when a question asks you to find ``the equilibrium'' the correct answer might involve multiple equilibria or no equilibria. The singular phrasing is used for brevity only, unless otherwise noted. 

    In discrete-type pricing problems you can assume that a customer that is indifferent between two deals will choose the one that gives the seller more profit. No need to mess around with ``minus one cent'' prices to handle tie-breaking.

    Sometimes it makes sense to use a computer for calculations, e.g., Excel, Mathematica, or Python/NumPy. If you do, still make sure to explain your reasoning and show the equations you solved.


    • Make sure to read the instructions at the "Assignments" page.

      10-27 14:30. Fixed a typo in 1.c.

    • 11-04 16:40.  Simplified 7.3b. It suffices to consider an equilibrium where neither company can benefit from adding one or discontinuing one existing flight (but no need to consider rescheduling existing flights). 

    • 11-14 17:05. Comment on Q3. In part a) the available bundling strategy is restricted to be pure, in parts b and c it is not restricted to be pure. In any part the profit-maximizing pricing scheme may or may not involve bundling. 

    • 11-17 10:10 Changed one numerical assumption in Q2 (maximum individual valuation is now $2, so the average valuation is $1). If you already answered the original version that is also fine, but it leads to a rather pathological (boring) case.

    • Notice the exceptional schedule. The final cut-off date is on Friday at noon, and early bonus is available until Thursday 15:00.