December 16, 2009, 14:00, MZG 8.136, Platz der Göttinger Sieben 5, 37073 Göttingen
Tertius de Wet, Stellenbosch University, South Africa
Semi-parametric estimation of inequality measures under heavy tailed distributions
Abstract:
Measures of inequality depend on the underlying distribution and their estimation then depend on the data available from such a distribution. In many cases such distributions have heavy-tails, either one-sided or two-sided, and care needs to be taken when estimating these measures since they usually include the tails of the distribution. Traditional estimators, using the empirical distribution function (edf), are typically not robust to data from heavy tailed distributions due to the fact that the data from the tails is often sparse compared to the centre of the distribution. This makes the edf non reliable in the tails and the inequality estimator non-robust. As an alternative, one can estimate the distribution in the centre using the edf and use a parametric estimator in the tails. Linking these together gives a semi-parametric estimator of the distribution function and using the latter in the estimation of measures of inequality, gives semi-parametric estimators for these measures. Since the parametric part is applied to the tails of the distribution, it follows naturally that use can be made of extreme value theory (EVT).
In this talk we will discuss the estimation problem in the above context and show how EVT can be used to obtain these semi-parametric estimators. Some indication is given of the planned extension of the research.
December 16, 2009, 15:00, MZG 8.136, Platz der Göttinger Sieben 5, 37073 Göttingen
Laurens de Haan, Erasmus University Rotterdam
Weak and Strong Financial Fragility
Abstract:
The stability of the financial system at higher loss levels is either characterized by asymptotic dependence or asymptotic independence. If asymptotically independent, the dependency, when present, eventually dies out completely at the more extreme quantiles, as in case of the multivariate normal distribution. Given that financial service firms’ equity returns depend linearly on the risk drivers, we show that the marginals’ distributions maximum domain of attraction determines the type of systemic (in-)stability. A scale for the amount of dependency at high loss levels is designed. This permits a characterization of systemic risk inherent to different financial network structures. The theory also suggests the functional form of the economically relevant limit copulas.
December 18, 2009, 11:15, seminar room IMS, Goldschmidtstr. 7
Dr. Zakhar Kabluchko, Universität Ulm
Limit theorems for sums of independent geometric Brownian motions
Abstract:
Let ξ1, ξ2,. . . . be independent copies of a standard Brownian motion ξ. Motivated by the results obtained previously in the context of the random energy model, we prove functional limit theorems for the process
N
ZN(t) = ∑ e ξi(sN + t)
i=1
as N →∞, where sN is a non-negative sequence converging to +∞. The limiting process depends heavily on the growth rate of the sequence sN. If sN grows slowly in the sense that lim infN → ∞ log N/sN > 2, then the limit is an Ornstein–Uhlenbeck process. However, if λ := limN → ∞ log N / sN Є (0, 2), then the limit is a certain completely asymmetric α-stable process Yα. We prove that the process Yα is stationary (α ≠ 1) and that it shares a number of properties of the Gaussian Ornstein – Uhlenbeck process.
January 8, 2010, 11:00, seminar room IMS, Goldschmidtstr. 7
Prof. Julie Le Gallo, Université de Franche-Comté, Besançon
Does manure management regulation work against agglomeration economies
Abstract:
A rise in geographical concentration of hog production and productivity has been observed from 1970s in USA and EU, suggesting the presence of agglomeration economies related to spatial spillovers and inter-dependencies among industries. In this paper, we determine whether manure management regulation restricting manure application per acre may weaken productivity gains arising from the agglomeration process. Our theoretical model shows that, on the one hand, dispersion is favored when manure is applied to land as a crop nutrient and, on the other hand, agglomeration is strengthened when manure treatment is adopted by farmers. Our estimations from French data suggest that the manure management regulation does not work against the spatial concentration of hog production but boosts the role played by non-market spatial externatlities in agglomeration process.
