Applying the Grey prediction model to the global integrated circuit industry

This study examines the precision of the Grey forecasting model applied to samples based on demand and sales in the global integrated circuit (IQ industry. In doing so, the main objective is to explore which forecast model is most appropriate for the IC industry by comparing the empirical results from the Grey model (GM), time series and exponential smoothing. Furthermore, three residual modification models are applied to enhance the forecasting results. Empirical results indicate that the GM is better suited to short-term predictions than to mid- and long-term predictions. Meanwhile, the Markov-chain residual modification model achieves reliable and precise results. (C) 2002 Elsevier Science Inc. All rights reserved.