On Determining the Dimension of Real-Time Stock-Price Data
利用Grassberger和Procaccia的相关积分法估计高频股票价格数据的维度，发现数据经滤波后维度较低，但使用延迟方法控制高阶矩依赖后，维度估计结果与随机过程类似，表明数据可能是低维高熵或非线性高维。
We estimate the dimension of high-frequency stock-price data using the correlation integral of Grassberger and Procaccia. The data, even after filtering, appear to be of low dimension. To control for dependence in higher moments, we use a new technique known as the method of delays in our reconstruction. Delaying the data leads dimension estimates similar to random processes. We conclude that the data are either of low dimension with high entropy or nonlinear but of high dimension.