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.