A modeling quality comparison of estimated Lévy models

Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F61989100%3A27510%2F10%3A10224832" target="_blank" >RIV/61989100:27510/10:10224832 - isvavai.cz</a>
Result on the web
—
DOI - Digital Object Identifier
—

Result language
angličtina
Original language name
A modeling quality comparison of estimated Lévy models
Original language description
The Lévy models, when used to simulate time series of returns, enable us to model kurtosis and skewness and thus overcome the main drawback of the Brownian motion. In this paper we focus on the two most widely processes from the Lévy's family of models,a variance gamma and a normal inverse Gaussian model. The variance gamma model can be regarded as a subordinated (geometric) Brownian motion driven by a random time with gamma distribution. In the normal inverse Gaussian model, Brownian motion is drivenby the inverse Gaussian distribution. Both of these models have four parameters, which need to be estimated. In the application part parameters for both models are estimated by means of a method of moments and a maximum likelihood method for five stock indices and five foreign exchange rates. Thereafter modeling quality of these estimated models is compared. Comparison is made on the basis of a log-likelihood function and errors of the basic descriptive statistics and quantile measures V
Czech name
—
Czech description
—

Project
<a href="/en/project/GA402%2F08%2F1237" target="_blank" >GA402/08/1237: Application of complex Lévy processes in modeling of financial assets prices</a><br>
Continuities
P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)<br>S - Specificky vyzkum na vysokych skolach

Publication year
2010
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Similar results(10)