Dynamic Local Trend Associations in Analysis of Comovements of Financial Time Series
We show that the correlation coefficient, often used for analysis of co-movements of financial time series, can be misleading because it does not take into account the time ordering of time series values. We propose the new method of analysis of time series comovements based on dynamic local trend association measure. This measure can capture the dynamic change of the sign of association between time series. The advantage of the new method is demonstrated on examples of financial time series. The associations between time series dynamics and related events are also considered.
KeywordsTime series Comovement Association measure Stock market Event Correlation
The work was supported by the projects IPN SIP 20171344 and SEP - CONACYT 283778, Mexico.
- 1.Batyrshin, I., Herrera-Avelar, R., Sheremetov, L., Panova, A.: Moving approximation transform and local trend associations in time series data bases. In: Perception-based Data Mining and Decision Making in Economics and Finance, pp. 55–83. Springer, Heidelberg (2007)Google Scholar
- 7.Peng, Y., Jiang, H.: Leverage financial news to predict stock price movements using word embeddings and deep neural networks. arXiv preprint arXiv:1506.07220 (2015)
- 12.Papadimitriou, S., Sun, J., Philip, S.Y.: Local correlation tracking in time series. In: Sixth International Conference on Data Mining (ICDM 2006), pp. 456–465. IEEE (2006)Google Scholar