On estimating the tail index and the spectral measure of multivariate \(\alpha \)-stable distributions
- 205 Downloads
We propose estimators for the tail index and the spectral measure of multivariate \(\alpha \)-stable distributions and derive their asymptotic properties. Simulation studies reveal the appropriateness of the estimators. Applications to financial data are also considered.
KeywordsAsymptotic distribution Multivariate α-stable distribution Spectral measure Tail index estimation Generalized empirical likelihood estimation
Mathematics Subject Classification60E07 62H12 62G32
The authors would like to thank an anonymous reviewer and Editor, Professor Norbert Henze, for their valuable comments and suggestions to improve the manuscript. They are also grateful to Professor John P. Nolan to provide access to STABLE package in R for improving the computations and Dr. Thaigo do Rego Sousa for his help in running the programs of GEVStableGarch package. The third author was supported by JSPS KAKENHI Grant Number 26870655.
- Fan Z (2001) Estimation problems for distributions with heavy tails. Ph.D. Thesis. Universität zu GöttingenGoogle Scholar
- Nolan JP (1999) Fitting data and assessing goodness of fit with stable distributions. In: Nolan JP, Swami A (eds) Proceedings of the conference on heavy tailed distributions. American University, WashingtonGoogle Scholar
- Nolan JP (2003) Modeling financial distributions with stable distributions. In: Rachev ST (ed) Handbook of heavy tailed distributions in finance. Elsevier Science, Amsterdam, pp 105–130Google Scholar
- Paulauskas V, Vaičiulis M (2011) Once more on comparison of tail index estimators. ArXiv:1104.1242v1
- Rachev ST, Stoyanov S, Biglova A, Fabozzi F (2005) An empirical examination of daily stock return distributions for U.S. stocks. In: Baier D, Decker R, Schmidt-Thieme L (eds) Data analysis and knowledge organization. Springer, Berlin, pp 269–281Google Scholar
- Rachev Z, Racheva-Iotova B, Stoyanov S (2010) Capturing fat tails. Risk magazine, pp 72–77. www.risk.net