Title: Využití Markovových řetězců pro predikování pohybu cen akcií
Other Titles: Using Markov chains for predicting the stock prices movement
Authors: Svoboda, Milan
Citation: Trendy v podnikání = Business trends : vědecký časopis Fakulty ekonomické ZČU v Plzni. 2014, č. 2, s. 63-70.
Issue Date: 2014
Publisher: Západočeská univerzita v Plzni
Document type: článek
article
URI: http://www.fek.zcu.cz/tvp/doc/akt/tvp-2-2014-clanek-8.pdf
http://hdl.handle.net/11025/16308
ISSN: 1805-0603
Keywords: analýza Markovových řetězců;přechod pravděpodobných matric;indikátory technické analýzy;obchodní strategie
Keywords in different language: Markov chain analysis;transition probability matrix;technical analysis indicators;trading strategies
Abstract in different language: This paper deals with stochastic modelling of short-term share prices movement using the Markov chains theory. The aim of the study is to find such prediction models which would be usable for creating business strategies for active trading at stock exchange. The research was carried out on the basis of day closing share prices of the company O2, negotiable at the Prague stock exchange during the seven years long period from the year 2006 to the year 2013. State space models were defined on the cumulative day price changes. On the basis of these cumulative day prices in total ten models of state space were calculated. We managed to find state space models with such states where a trend change occurs with sufficiently high probability. On four of these models the business strategies were applied. There were16 business strategies applied on each model. In most of these strategies the passive buy-and-hold strategy was beaten. The highly valued aspect for the further research is the fact that it was a compact set of winning models and business strategies.
Rights: © Západočeská univerzita v Plzni
Appears in Collections:Číslo 2 (2014)
Články / Articles (KEM)
Číslo 2 (2014)

Files in This Item:
File Description SizeFormat 
Svoboda.pdfPlný text532,54 kBAdobe PDFView/Open


Please use this identifier to cite or link to this item: http://hdl.handle.net/11025/16308

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.