Generalized Hyperbolic Distributions: Empirical Evidence on Bucharest Stock Exchange

Olivia Andreea Baciu

back

Abstract:

Onfive of the most liquid and important equities of the Romanian stock market together with the market index is investigated the fit of the generalized hyperbolic distributions. The parameters of the hyperbolic distribution, Variance- Gamma, Normal Inverse Gaussian, skewed t Student and generalized hyperbolic are estimated using the maximum likelihood estimation. The goodness-of-fit measures used to assess the fitofeachdistribution are the Kolmogorov- Smirnov distance, Akaike information criteria and the log- likelihood. Plots are also inspected. The Variance- Gamma distribution was ruled out by the Kolmogorov- Smirnov test. After inspecting the plots, a good approximation of the data was given by the Normal Inverse Gaussian distribution and the generalized hyperbolic, but based on the goodness-of-fit measures, the generalized hyperbolic distribution yield to be the best fit.

 

 

 

 
Copyright © 2009 | All rights reserved