Modeling candlestick patterns with interpolative boolean algebra for investment decision making

Abstract

In this paper we present one way of modeling candlestick patterns using interpolative Boolean algebra (IBA). This method shows a degree of fulfillment for observed patterns, thus giving traders easy interpretation of by how much candlesticks fit into different patterns. Candlestick patterns have been used for financial forecasting for a couple of decades on Western markets and they have become a mainstream trader's tool. Since the need for automated candlestick patterns discovery arose, some papers proposed fuzzy approach as a solution. Our decision to use IBA for modeling candlestick patterns comes from the fact that fuzzy logic has it limits and cannot be applied to these models. Proposed method is another approach to the same problem, but it could not be modeled using conventional fuzzy logic, because it is necessary for it to be in the Boolean frame. Results obtained from our tests are satisfactory and also open the opportunity for combining this technique with existing ones. © 2013 Springer-Verlag Berlin Heidelberg.