- 期刊
Innovative Prediction Method: Modified High-order Multi-dimension Non-equidistant Grey Model GM(1,1)
並列摘要
Because the traditional Grey Model is not very efficient in large, change data, hence, GM hasn’t been extensively applied to financial field. In order to improve the accuracy of predicting the large change data, we modified the traditional high-order multi-dimension non-equidistant GM(n,m) and denote this modified model by mGM(n,m). Combining with empirical mode decomposition and modified Savitaky-Golay filter, we obtain this modified method based on one-dimension modified mod- el mGM(n,1). Besides, we use mGM (2,m) to forecast the closing price and the turnover rate. From both the rational mathematical derivative process and the actual predicted result, we have sound reasons to believe that this modified grey model is efficient for the large change data.
延伸閱讀
- 王凱立、林嘉慧（2003）。A New Parametric Approach to Modeling Generalized Autoregressive Conditional Density Model at Higher Order Moments。財務金融學刊，11(2)，1-42。https://doi.org/10.6545/JFS.2003.11(2).1
- Wang, F. K., Hsiao, Y. Y., & Chang, K. K. (2012). Combining Diffusion and Grey Models Based on Evolutionary Optimization Algorithms to Forecast Motherboard Shipments. Mathematical Problems in Engineering, 2012(), 146-155. https://doi.org/10.1155/2012/849634
- Chen, S. P., & Shih, C. M. (2007). Diffusion Forecasting of Innovative Products Using an Improved Grey Model. Journal of Grey System, 10(1), 23-32. https://doi.org/10.30016/JGS.200706.0004
- Huang, Q. F., & Wei, Y. (2010). Grey Forecasts of Combination Model-GM(2,1) and Time-series. Journal of Grey System, 13(4), 127-132. https://doi.org/10.30016/JGS.201012.0001
- 王佩淳（2004）。Combining Grey Theory and Quality Function Deployment to Construct a Dynamic Trend Prediction Model〔碩士論文，亞洲大學〕。華藝線上圖書館。https://www.airitilibrary.com/Article/Detail?DocID=U0118-0807200916284190