The recent upswing in the popularity of economic applications of `networks' may be attributed to the relative e ciency of network theory in analyzing complex economic structures with much more clarity compared to any other (standard) methods used by economists. Having said that, it can hardly be denied that most of the network-based economic models su er from lack of availability of `ready-to-use' network datasets. As such, a number of times, researchers are forced to base their analyses on simulations or pick from the small number of datasets available (for example, ICPSR Add Health Dataset). This tends to limit the variety of empirical exercises that could be carried out to understand some economic networks better. For instance, consider an interbank network. Most of the datasets with public access do not have information about the bilateral connections between banks within this network. Under such circumstances, conducting any form of predictive or explanatory analyses with regard to the manner in which connections are formed in such a network is almost impossible. Upper and Worms (2004) addressed this issue by using the technique of Maximum Entropy (ME) to estimate these connections from the (publicly available) information on total assets and liabilities of the banks in the German interbank network. However, as will be discussed shortly, ME has been proven to have some obvious shortcomings. This serves as the motivation for our current research. In this paper, we construct a `copula' based approach instead, and establish its merits over the Maximum Entropy approach. More than an all purpose methodology, our paper aims to extend the toolkit of nancial stability evaluation that allows an analyst to make di erent ∗Department of Economics, Indiana University, Bloomington, IN 47408 †Faculdade de Economia da Universidade do Porto, Porto, Portugal This research is, currently, in progress. The authors expect it to be completed in time for presentation at the `Networks in Trade and Finance' conference to be held in November , 2012.
[1]
Ethan Cohen-Cole,et al.
Systemic Risk and Network Formation in the Interbank Market
,
2010
.
[2]
Ben R. Craig,et al.
Interbank Tiering and Money Center Banks
,
2010
.
[3]
Leibniz-Informationszentrum Wirtschaft,et al.
Estimating Bilateral Exposures in the German Interbank Market: Is there a Danger of Contagion?
,
2002
.
[4]
João F. Cocco,et al.
Lending Relationships in the Interbank Market
,
2003
.
[5]
Simon J. Wells,et al.
Financial Interlinkages in the United Kingdom's Interbank Market and the Risk of Contagion
,
2004
.
[6]
Craig H. Furfine,et al.
Interbank Exposures: Quantifying the Risk of Contagion
,
1999
.
[7]
Michael Boss,et al.
Network topology of the interbank market
,
2003,
cond-mat/0309582.
[8]
G. Caldarelli,et al.
A Network Analysis of the Italian Overnight Money Market
,
2005
.
[9]
Stavros A. Zenios,et al.
A Comparative Study of Algorithms for Matrix Balancing
,
1990,
Oper. Res..
[10]
G. Iori,et al.
Trading strategies in the Italian interbank market
,
2006,
physics/0611023.