Statistical and NN Forecasting Models of Financial Data: Making Inferences about the Accuracy and Risk Evaluation
Milan Marcek, Jindrich Petrucha and Dusan Marcek
For both accurate and easy risk consideration in decision making, it is necessary to have flexible and soft methods that adapt to the input data and problem conditions. In this paper, we consider the accuracy of forecasting models based on statistical (stochastic) methods sometimes called hard computing and soft methodology based on the soft or granular computing. A new method for finding the forecasting horizon within which the risk is minimal is also presented. To evaluate the risk we used methods based on the analysis of forecast errors by applying the exponential smoothing concept. It is also found that the risk estimation process based on soft methods is simplified and less critical to the question whether the data is true crisp or white noise.
Keywords: Risk in decision making, forecasting accuracy, time series, ARCHGARCH models, soft neural networks, granular computing.