Parts | Days | Selection | Search | Downloads | Help
SOE: Fachverband Physik sozio-ökonomischer Systeme
SOE 21: Financial Markets and Risk Management II
SOE 21.2: Talk
Thursday, March 25, 2010, 10:45–11:15, H46
Optimal estimation of power laws with applications to socioeconomic data — •Faustino Prieto and Jose Maria Sarabia — University of Cantabria, Santander, Spain
Power laws appear widely in many branches of economics, finance, physics, computer science, demography and social sciences. The upper tail of many sets of data, including the size of cities, personal income, earthquakes, forest fires and many other examples all appear to follow power laws. A crucial point in the estimation of these laws is the correct choice of the truncation point. The aim of this paper is to investigate how to choice this truncation point from an optimal point of view. A new methodology based on the Akaike information criterion is proposed. An extensive simulation study is carried out in order to prove the existence of this optimal point, under different assumptions about the underlying population. Several kinds of populations are considered, including lognormal and populations with heavy tails. Finally, the methodology is used to optimal estimation of power laws in socioeconomic data sets, including city and business size data.