集成模型ARIMAX-GARCH及其在股票预测中的应用

Integrated Model ARIMAX GARCH and Its Applications in Stock Forecast

在分析ARIMAX模型与GARCH模型的预测特性和优劣的基础上,建立了基于两者集成的ARIMAX-GARCH模型,其基本思想是充分发挥两种模型在回归与序列波动性因素提取方面的优势.对从5大行业中随机抽取的10只只股票的实证分析表明,该集成模型在股票预测中的准确率与稳定性显著优于两个单一模型.

Based on the analysis of the performance of multiple stationary time series of ARIMAX and GARCH models, this paper sets up a new model which integrates ARIMAX with GARCH. The new model has the advantage of regression in ARIMAX and the superiority of extracting volatility in GARCH. The result of empirical analysis about ten shares from five industries which are grabbed at random shows that the proposed model has better accuracy and stability in stock forecast than every single model.