Adaptative Holt’s forecasting model based on immune paradigm

Tomasz Pelech


In modern management, decision making causes necessity of forecasting (with appropriate methods) to reach profitable and smooth functioning of companies as well as of small and medium businesses. The Holt’s model is a method that can be used to predict exchange quotations. It is based on values of two parameters that estimate expected value and its trend. To construct dynamic forecasting model it is necessary to apply adaptive rules of the parameters adjusting. The paper shows applicability of an immune-based paradigm on the example of adaptive Holt’s model. Presented results of computations with the constructed algorithm show that immune paradigm can be thought as promising way to improve results (accuracy) of forecasting. They are not absolutely better than the results acquired with the classical Holt’s method because of some factors like the length of input data, characteristic of the method and construction of algorithm that is still developed.

Полный текст:

PDF (English)


. de Castro L.L., von Zuben F.J.: «Artificial Immune Systems: Part I – Basic Theory and Applications»,

. Holt C.C.: «Forecasting seasonals and trends by exponentially weighted moving averages», Pittsburgh, Pennsylvania, Carnegie Institute of Technology, 1957.


  • На текущий момент ссылки отсутствуют.