|
on Discrete Choice Models |
| By: | Atila Abdulkadiroglu (Columbia University); Parag A. Pathak (Harvard University); Alvin E. Roth (Harvard University); Tayfun Sönmez (Boston College) |
| Abstract: | In July 2005 the Boston School Committee voted to replace the existing Boston school choice mechanism with a deferred acceptance mechanism that simplifies the strategic choices facing parents. This paper presents the empirical case against the previous Boston mechanism, a priority matching mechanism, and the case in favor of the change to a strategy-proof mechanism. Using detailed records on student choices and assignments, we present evidence both of sophisticated strategic behavior among some parents, and of unsophisticated strategic behavior by others. We find evidence that some parents pay close attention to the capacity constraints of different schools, while others appear not to. In particular, we show that many unassigned students could have been assigned to one of their stated choices with a different strategy under the current mechanism. This interaction between sophisticated and unsophisticated players identifies a new rationale for strategy-proof mechanisms based on fairness, and was a critical argument in Boston's decision to change the mechanism. We then discuss the considerations that led to the adoption of a deferred acceptance mechanism as opposed to the (also strategy-proof) top trading cycles mechanism. |
| Date: | 2006–01–07 |
| URL: | http://d.repec.org/n?u=RePEc:boc:bocoec:639&r=dcm |
| By: | Christos Genakos |
| Abstract: | This paper examines how information on the purchasing patterns of differentcustomer segments can be used to more accurately evaluate the economicimpact of mergers. Using a detailed dataset for the leading manufacturers in theUS during the late nineties, I evaluate the welfare effects of the biggest ($25billion) merger in the history of the PC industry between Hewlett-Packard andCompaq. I follow a two-step empirical strategy. In the first step, I estimate ademand system employing a random coefficients discrete choice model. In thesecond step, I simulate the postmerger oligopolistic equilibrium and compute thewelfare effects. I extend previous research by analysing the merger effects notonly for the whole market but also for three customer segments (home, smallbusiness and large business). Results from the demand estimation and mergeranalysis reveal that: (i) the random coefficients model provides a more realisticmarket picture than simpler models, (ii) despite being the world's second andthird largest PC manufacturers, the merged HP-Compaq entity would not raisepostmerger prices significantly, (iii) there is considerable heterogeneity inpreferences across segments that persists over time, and (iv) the merger effectsdiffer considerably across segments. |
| Keywords: | Computer industry, discrete choice models, merger analysis, productdifferentiation, random coefficients. |
| JEL: | D12 G34 L41 L63 |
| Date: | 2004–12 |
| URL: | http://d.repec.org/n?u=RePEc:cep:stieip:39&r=dcm |
| By: | Patrick Bajari; Han Hong; John Krainer; Denis Nekipelov |
| Abstract: | We propose a method for estimating static games of incomplete information. A static game is a generalization of a discrete choice model, such as a multinomial logit or probit, which allows the actions of a group of agents to be interdependent. Unlike most earlier work, the method we propose is semiparametric and does not require the covariates to lie in a discrete set. While the estimator we propose is quite flexible, we demonstrate that in most cases it can be easily implemented using standard statistical packages such as STATA. We also propose an algorithm for simulating the model which finds all equilibria to the game. As an application of our estimator, we study recommendations for high technology stocks between 1998-2003. We find that strategic motives, typically ignored in the empirical literature, appear to be an important consideration in the recommendations submitted by equity analysts. |
| JEL: | L0 L5 C1 |
| Date: | 2006–02 |
| URL: | http://d.repec.org/n?u=RePEc:nbr:nberwo:12013&r=dcm |