从统计套利看大数据的研究与应用

大数据是一个热门词,但还没有形成严格的理论基础。研究大数据的目的全在于应用。从数学形式来看,大数据与高维高频海量数据区别不大.从统计学的观点来看,研究大数据就是从高维高频的海量数据中找出一个较低维的平稳过程,然后利用大数定律(也叫遍历定理)找到其可用的价值。在金融交易中,这就是统计套利.

From statistical arbitrage to big data study

“Big data” is a hot term. However its strict theoretical basis has not been formed yet.The purpose of research in big data is from its application. From their mathematical formulation,big data and massive high-dimensional and high-frequency data have no big difference. From the statistical point of view, the big data study is to find a lower-dimensional stationary time series from the high-dimensional and high-frequency massive data, then use the law of large numbers(the ergodic theorem) to find the value of its application. In finance, that is just statistical arbitrage.