SSLPNet: A financial econometric prediction model for small-sample long panel data

Abstract

The recent period has witnessed the quantification and modeling of financial data become the crystallization of the intersection of finance and computers. In some research institutions, this crystallization product has received a specific name - FinTech. Numerous studies based on big data processing tend to obtain indicators with very optimistic accuracy, while for long panel-type small sample data, the existing studies propose very sparse targeted models. In this paper, we analyze the trend of trading volume and price move-ment of 10 stocks based on tick frequency data of 10 stocks in cross-section time series and set up a split-order algorithm to obtain the maximum total trading volume under the condition of satisfying the predicted trend to assist investors to maximize their