Kalman Filter to Estimate Dynamic and Important Patterns of Interaction between Multiple Variables pp. 289-320
Authors: (Harya Widiputra, Russel Pears, Nikola Kasabov, The Knowledge Engineering and Discovery Research Institute, Auckland University of Technology, New Zealand)
Abstract: Estimating the state of nature processes has been a very challenging task for human beings. These processes include not only those which occur in the biological field, i.e., interaction between genes; or those which emerge in the ecological field, i.e., how humidity is related to the level of sun ray, wind speed and rain; but also processes which exist in the global financial area, i.e., how a stock market in a specific country influences or is being influenced by the other stock markets in different countries directly or indirectly, and how the macro economic factors in a single country are related to each other from time to time. Being able to model past and present states of these processes would lead us to the possibility of understanding how various variables or things in this world interact, which in the end would offer us the knowledge to estimate future states of the systems in which these processes take place.