Adaptative Holt’s forecasting model based on immune paradigm

Abstract

In modern management, decision making necessitates forecasting (using appropriate methods) to achieve profitable and smooth functioning of both companies and SMEs. The Holt model is a method that can be used to forecast stock exchange quotes. It is based on the values of two parameters estimating the expected value and its trend. To build a dynamic forecasting model, it is necessary to apply adaptive rules for adjusting the parameters. The paper shows the applicability of the immune paradigm on the example of Holt's adaptive model. Preliminary results of calculations using the constructed algorithm show that the immune paradigm can be considered as a promising way to improve the results (accuracy) of forecasting. They are not absolutely better than the results obtained with the classical Holt method due to a number of factors such as the length of input data, the peculiarity of the method and the design of the algorithm, which is still in the development stage.