[论文解读] Impact of Data Normalization on Deep Neural Network for Time Series Forecasting

Samit Bhanja, Abhishek Das|arXiv (Cornell University)|Dec 13, 2018

Stock Market Forecasting Methods参考文献 14被引用 102

一句话总结

本文研究了各种数据归一化技术如何影响深度递归神经网络在时间序列预测中的表现，并将 DRNN 应用于 BSE 和 NYSE 收盘指数。

ABSTRACT

For the last few years it has been observed that the Deep Neural Networks (DNNs) has achieved an excellent success in image classification, speech recognition. But DNNs are suffer great deal of challenges for time series forecasting because most of the time series data are nonlinear in nature and highly dynamic in behaviour. The time series forecasting has a great impact on our socio-economic environment. Hence, to deal with these challenges its need to be redefined the DNN model and keeping this in mind, data pre-processing, network architecture and network parameters are need to be consider before feeding the data into DNN models. Data normalization is the basic data pre-processing technique form which learning is to be done. The effectiveness of time series forecasting is heavily depend on the data normalization technique. In this paper, different normalization methods are used on time series data before feeding the data into the DNN model and we try to find out the impact of each normalization technique on DNN to forecast the time series. Here the Deep Recurrent Neural Network (DRNN) is used to predict the closing index of Bombay Stock Exchange (BSE) and New York Stock Exchange (NYSE) by using BSE and NYSE time series data.

研究动机与目标

• 推动重新定义用于非线性、动态时间序列预测的 DNN 模型的必要性。
• 评估不同数据归一化技术对学习和预测性能的影响。
• 在将数据输入 DRNN 模型之前，评估预处理、网络架构和参数选择。

提出的方法

• 在将时间序列数据输入 DRNN 之前应用不同的归一化方法。
• 使用深度递归神经网络预测 BSE 和 NYSE 的收盘指数。
• 比较归一化方法对 DRNN 预测结果的影响。

实验结果

研究问题

• RQ1 不同的数据归一化技术如何影响基于 DRNN 的时间序列预测？
• RQ2 哪种归一化方法为 BSE 与 NYSE 数据提供最有效的预测性能？
• RQ3 数据归一化在塑造 DRNN 对非线性时间序列的学习动力学中的作用是什么？

主要发现

• 本文探讨了归一化技术对 DRNN 在时间序列预测中的表现的影响。
• 摘要片段中未提供具体数值结果。