. . "23510" . . "Jihlava" . . "Application of non-homogeneous Markov chain analysis to trend prediction of stock indices" . . "Jihlava" . "[8EA880ED8193]" . . "Markov chain analysis, transition probability matrix, stock index, trend prediction, time series analysis."@en . "Application of non-homogeneous Markov chain analysis to trend prediction of stock indices"@en . . . "2"^^ . "RIV/49777513:23510/13:43919711!RIV14-MSM-23510___" . "978-80-87035-76-4" . "2"^^ . "Application of non-homogeneous Markov chain analysis to trend prediction of stock indices" . . "Svoboda, Milan" . . "The present paper concerns with prolongation of our study aims to predict the stock index trend of various stock indices using Markov chain analysis (MCA). The prediction of the trend using MCA is done using short, medium and long-term data. Each downloaded data is divided into two periods. The first one is used for estimating the trend, whereas the second one is used for comparison and evaluation, too. In the basic framework, we may use either homogeneous MC or nonhomogeneous one. When building model we are focused both on various state space discretizations and corresponding construction of transition probability matrices. Non-homogeneous matrices are constructed by linear interpolants between two transition probability matrices at given time steps. These objects represent a core of any MCA and its application. The results of the trend prediction using both versions of MCA are compared. Numerical calculations and computer implementations have been done by Excel and Mathematica modules, which are briefly discussed as well." . . . "61785" . "S" . "6"^^ . . . "College of Polytechnics Jihlava" . "2013-09-11+02:00"^^ . . "Luk\u00E1\u0161, Ladislav" . "The present paper concerns with prolongation of our study aims to predict the stock index trend of various stock indices using Markov chain analysis (MCA). The prediction of the trend using MCA is done using short, medium and long-term data. Each downloaded data is divided into two periods. The first one is used for estimating the trend, whereas the second one is used for comparison and evaluation, too. In the basic framework, we may use either homogeneous MC or nonhomogeneous one. When building model we are focused both on various state space discretizations and corresponding construction of transition probability matrices. Non-homogeneous matrices are constructed by linear interpolants between two transition probability matrices at given time steps. These objects represent a core of any MCA and its application. The results of the trend prediction using both versions of MCA are compared. Numerical calculations and computer implementations have been done by Excel and Mathematica modules, which are briefly discussed as well."@en . "RIV/49777513:23510/13:43919711" . "Mathematical Methods in Economics 2013: 31st International Conference: Proceedings" . "Application of non-homogeneous Markov chain analysis to trend prediction of stock indices"@en . . . . .