In this part of the paper we deal with defining a prediction model which determines the future val-ues of the state on the base of a given sequence of its past values (time series). The prediction model has a form of approximation of the searched functional relationship by a polynomial of first degree (affine mapping), or by a polynomial of second degree (quadratic). The so called local model of approximation has been discussed and applied. The computations concern the data of 10 days walking average run-off the Wupper Reservoir Water System in Germany. The run-off process has the character of deterministic chaos. The respective computer programs have been written in Fortran 90. They determine the values predicted for up to three steps in the time series. The evaluation of the prediction quality, by some statistical criteria, is presented.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.