1M0572 GA MŠMT - Ministry of Education, Youth and Sports (MEYS)
GA201/02/1391 GA ČR - Czech Science Foundation (CSF)
IAA1075403 GA AV ČR - Academy of Sciences of the Czech Republic (AV ČR)
CEZ
AV0Z10750506 - UTIA-B (2005-2011)
Annotation
Errors of the Neyman-Pearson testing and risks of the Bayesian decisions are estimated by means of more easily evaluated information-theoretic divergences when the hypothesis is the geometric Brownian motion and the alternative is a diffusion random process with price-dependent growth rate. A series of mathematical theorems is proved in order to characterize the properties of the obtained estimates. Concrete DAX-index type data are used to illustrate the accuracy and practical value of the estimates.