Trend analysis with use of binary representation
AbstractOne of the basic methods of technical analysis is the trend analysis. There are many indicators that allow for a current trend detection and parameter specification based on the candlestick chart representation. Because the use of candlestick representation is characterized by lower informative value, i.e. leads to a loss of information about the course trajectory, better modelling effects can be obtained by using a binary representation. The binarization algorithm transforms the course represented by tick data into a corresponding binary sequence. In the article are proposed algorithm allowing for a detection of current trend and specification of its parameters using binary representation. As a result of the algorithm performance we obtain, for each change in the binary representation, the type of the current trend and its parameters, resulting in a so-called binary-trend representation of a course trajectory. In the article we also include an exemplary statistical analysis of trends, performed based on historical tick data from a six-year period (2013-2019), for AUD/NZD currency pair. In order to execute the research a dedicated software was created in MQL4 and C++.
|Publication size in sheets||0.5|
|Book||Houda Michal, Remeš Radim (eds.): 37th International Conference Mathematical Methods in Economics MME 2019 : Conference Proceedings, 2019, University of South Bohemia in České Budějovice, Faculty of Economics, ISBN 978-80-7394-760-6, 598 p.|
|Keywords in Polish||rynek walutowy, ekonometria wysokich częstotliwości, analiza techniczna, analiza trendum modelowanie kursuów watlutowych, wspomaganie decyzji inwestycyjnych|
|Keywords in English||foreign exchange market, high frequency econometric, technical analysis, trend analysis, modelling of currency exchange rates, currency market investment decision support|
|Score||= 5.0, 30-01-2020, ChapterFromConference|
* presented citation count is obtained through Internet information analysis and it is close to the number calculated by the Publish or Perish system.