Reconstruction of financial time series data based on compressed sensing

Time series data are widely used in financial research; however, data frequency and completeness can greatly affect the research results. Although high-frequency financial time series data can be obtained, some scenarios, such as bank lending data, may lack high frequency. Currently, mainstream data interpolation methods should improve the data reconstruction accuracy. In this study, we improve the compressed sensing method to expand its field of application, specifically for reconstructing financial data. The results show that the data reconstruction based on compressed sensing can effectively improve the reconstruction accuracy.

Automated summary

Time series data play a crucial role in financial research, yet data frequency and completeness significantly affect the research outcomes. This study enhances the compressed sensing method for reconstructing financial data and demonstrates its effectiveness in improving reconstruction accuracy.