论文部分内容阅读
在金融企业中,时间序列是一种重要的数据类型。高效、准确地预测金融时间序列对于企业的运作具有重要意义。提出使用一种具有增量学习能力的模糊神经网络(FNN-IL)应用于金融时间序列的预测。FNN-IL能学习蕴涵在时间序列中的知识,并能跟踪时间序列的运行从而动态调整模糊规则库。对比试验表明FNN-IL的性能优于传统的FNN。
In financial enterprises, time series is an important data type. Efficient and accurate prediction of financial time series is of great significance to the operation of enterprises. A fuzzy neural network with incremental learning ability (FNN-IL) is proposed to predict financial time series. FNN-IL can learn the knowledge implied in the time series, and can track the running of the time series to dynamically adjust the fuzzy rule base. Comparative tests show that FNN-IL performs better than traditional FNN.