Algorithmic Trading Using Markov Chains: Comparing Empirical and Theoretical Yields
Algorithmic Trading Using Markov Chains: Comparing Empirical and Theoretical Yields

Author(s): Milan Svoboda, Pavla Říhová
Subject(s): Methodology and research technology, Policy, planning, forecast and speculation, Financial Markets
Published by: Masarykova univerzita nakladatelství
Keywords: algorithmic trading; stock market predication; Markov chains analysis;
Summary/Abstract: This study is focused on comparing empirical and theoretical yield of business strategies applied to stock markets. We continue in our previous articles in which we deal with the short-term prediction of stock markets and with creating business strategies using Markov Chains analysis. When defining a state space we assume that the stock price moves in three types of trends: primary, secondary and minor. The object of our interest is a minor trend which usually lasts for several days. During this trend the stock price accumulates a certain profit or loss in relation to the price at the beginning of the trend. The state space is defined by the size of the accumulated profit or loss. Business strategies are formed in the way that the states in which the stock price decreases generate buying signals and the states in which the stock price increases generate selling signals. Theoretical profitability of a business strategy is modeled on the basis of a matrix of transition between states probability, a matrix of evaluation of these transitions and the expected number of transactions. We calculate the parameters of this model, as well as empirical profitability, with historical data of CEZ stock during the ten years period from early 2006 to the end of 2015. In some cases empirical and theoretical results were nearly the same, in other cases they differed significantly.

• Details
• Contents