CAJ | 학술논문

中国股指期货的推出指日可待，交易者多了一种投资工具的同时也带来了新的风险。建立准确的金融时间序列预测模型是逐利及避险的方法之一，一直是学者专家研究的热点。本研究结合小波转换与支持向量回归，提出一个二阶段时间序列预测模型。先以离散小波框架将预测变量分解成不同尺度的多个子序列，揭示隐藏在预测变量内的信息，再以支持向量回归为工具，以这些子序列为预测变量建构SVR模型。本研究以日经225指数开盘价为预测目标，以期货开盘价为预测变量对模型进行实证研究，结果显示，该模型的预测绩效比单纯SVR模型及随机漫步模型好。未来可尝试以不同的基底函数作进一步研究。
중국고지기화적추출지일가대，교역자다료일충투자공구적동시야대래료신적풍험。건립준학적금융시간서렬예측모형시축리급피험적방법지일，일직시학자전가연구적열점。본연구결합소파전환여지지향량회귀，제출일개이계단시간서렬예측모형。선이리산소파광가장예측변량분해성불동척도적다개자서렬，게시은장재예측변량내적신식，재이지지향량회귀위공구，이저사자서렬위예측변량건구SVR모형。본연구이일경225지수개반개위예측목표，이기화개반개위예측변량대모형진행실증연구，결과현시，해모형적예측적효비단순SVR모형급수궤만보모형호。미래가상시이불동적기저함수작진일보연구。
It＇s an important method for profiting and risk-evading to construct financial time series forecasting model. This paper brings forward a two stage model combining with wavelet transform and support vector regression to predict the stock price, and uses the wavelet frame to decompose the predict variable to be several subseries with different scales to make the information hidden in predict variable known; and then constructs a SVR forecasting model with these subseries. The purpose is to improve the forecasting accuracy of SVR model through pre-process the predict variable by wavelet frame. In order to validate the forecasting capability, this paper designs an empirical research using the futures price of Nikkei 225 to predict the Nikkei 225 spots open price based on the two stage model. The empirical results show that the proposed model outperforms the SVR model and random walk model, the cumulating returns from the stratagem proposed by this two-stage model is also better than from other models.