Placement Director: Vijay Krishna
    (814) 863-8543
    vkrishna@psu.edu

Contact Information:
Avantika Chowdhury
  Office: (814) 865-2745
  Cell: (814) 777-2487
E-mail: avantikac@psu.edu
Website: www.econ.psu.edu/~auc134

CITIZENSHIP:

• India (F-1 Visa)

EDUCATION:

• Ph.D., Economics, Penn State University, expected May 2008.
• M.S., Economics, Indian Statistical Institute, Delhi, May 2003.
• B.Sc. Economics, Calcutta University, August 2001.

PH.D. THESIS:

• “Essays on the microeconomics of network formation” Thesis Advisor: Professor Kalyan Chatterjee

FIELDS:

• Primary: Microeconomic Theory, Game Theory
• Secondary : Industrial Organization, Development Economics

PAPERS:

• “Academic Citations and Diffusion of Knowledge:  An  Economic Analysis” with K. Chatterjee, February 2007.
• “Competing to be a ‘Star': A model of sequential network formation” November 2007

GRANTS &
FELLOWSHIPS:

• David. W. and Carolyn. P. Grow Graduate Fellowship, 2005
• College of Liberal Arts Dissertation Support Grant, Fall 2007

TEACHING EXPERIENCE:

• Instructor: Intermediate Microeconomics (1 semester)
• Teaching Assistant: Intermediate Microeconomics (3 semesters), Economics of Corporation (2 semesters), Game Theory (PhD level, 3 semesters)

RESEARCH EXPERIENCE:

• Research Assistant to Kalyan Chatterjee, Summer 2007 and Summer 2005

PRESENTATIONS & OTHER PROFESSIONAL ACTIVITIES:

• "Competing to be a ‘Star': A model of sequential network formation," Cornell-PSU Workshop, September 2007

REFERENCES:

• Professor Kalyan Chatterjee
• Professor Neil Wallace
• Professor Marek Pycia

THESIS ABSTRACT

In this paper, we model network formation as a sequential move game among agents who have perfect foresight and are aware of the consequence of their decisions on the future. This is in contrast with most of the existing literature that either considers a single round of simultaneous link formation or have myopic decision-makers. In the basic model, players are homogeneous and each can form only one link (an extreme form of capacity constraint). The payoff from being connected to another player in a network diminishes (decays) geometrically with the distance between them. This decay factor drives a wedge between the payoffs of different agents depending on the centrality of their positions in the network. The sequential move structure enables us to consider the tradeoff between the desire of agents to connect to the currently most central player and to become the most central player in the network himself or herself.

We find that the equilibrium network structure depends crucially on the rate of decay. When the rate of decay with distance is low, the subgame perfect equilibrium structure is a single star. When it is high, the equilibrium structure is a combination of a wheel and a star but still connected. For intermediate ranges of this rate, however, equilibrium could entail more than one connected component leading to formation of groups of different sizes. Equilibrium network structures, in general, are not efficient except for the case of a very low rate of decay. With an increase in capacity, the networks for a very low and a very high level of decay are, respectively, a star with multiple centres and a generalized wheel with multiple local stars.

Essay 2. “Academic Citations and Diffusion of Knowledge: An Economic Analysis” (joint with Kalyan Chatterjee)

There has been recent empirical work on the structure of citation networks in specific academic fields, following on the celebrated early work of Price and Merton. It is shown that most of these networks, like other real-world networks, exhibit preferential attachment, that is, there are a few nodes with many connections and several with none or very few. The models used to explain this have concentrated on random link formation and on copying, without considering the fact that academic citations are often used as a measure of academic success and therefore have implications for academic careers. We construct a model in which competition for citations plays a central role and show that this could slow down the rate of diffusion of useful knowledge, depending on the behavioural rule used when a new entrant is indifferent among citing several previous entrants. Adding private information about quality and social acquaintance as a factor in citation, we are able to generate something that resembles the preferential attachment models of the empirical literature. We also consider strategic entry of authors in a given field and show that this results in a cascade. Though we might have ex-ante efficiency in some equilibria, ex-post efficiency is not guaranteed since it is always possible, in equilibrium, that a good paper “dies” and a worse one survives.