Problems - Ch9

  1. Create the following simulated data.

    1. x1 ~ U[0,1], x2 ~ U[0,1], e ~ N(0,1)

    2. v = a + b*x1 + c*x2 + e, a=-3,b=2, c=4

    3. Num ~ 2 + round(10*U[0,1])

    4. For each of 200 auctions

      1. Draw number of bidders (Num)

      2. Draw x1 and x2 (characteristics of the auction)

      3. Simulating bids for Num bidders (assume b = v)

      4. Set the price equal to the second highest bid

      5. Create a data set with auction id, number of bidders, x1, x2, price.

      6. Present summary statistics

    5. Run ordinary least squares on the relationship between price, x1 and x2 (price as a function of x’s). Present estimates for a, b and c. Discuss why they are or are not close to the true values.

    6. Run ordinary least squares on the relationship between price, x1, x2 and number of bidders. Present estimates for a, b and c. Discuss why they are or are not closer to the true values than for (5).

  2. Create a non-parametric English auction estimator and estimate the parameters using the data created in (1).

  3. Download the data for A Study of the Internal Organization of a Bidding Cartel. The data is available here: http://www.johnasker.com/DataSampleStamps.xls

      1. Replicate Table 3 columns (1) OLS, to the best you can. Note the data you have is only a sample of the full data used in the paper.

      2. Economic theory states that bidders in a simple sealed bid auction should bid more when there are more competitors. In the auction here it is less clear what the relationship is. In these knockout auctions, do bidders increase their bid as the number participants increase? Using one or more of the methods, estimate the relationship between the bid amount and the number of competitors in the knockout auction.

      3. Explain the concerns you have with the estimate in (2).

      4. Suggest alternative methods or data that would alleviate your concerns raised in (3).