Title: Statistical problems in Markov chains with applications in finance Author: Marek Chudý
Department: Department of Probability and Mathematical Statistics Supervisor: RNDr. Michaela Prokešová, Ph.D.
Abstract: In this work, we study estimation methods for estimating transition probabilities in Markov chains. We discuss two methods, the first one for com- plete data and the second one for aggregate data. In the second chapter, we will introduce the theory for both methods and show examples of tests of sev- eral hypothesis about transition probabilities. In the last chapter we apply both methods to real data comming from an insurance company. In the last chapter we also present the results of both methods and compare them with each other.
Keywords: Markov chains, transition probabilities, maximum likelihood method, least squares method