Data for: Application of asymmetric proximal support vector regression based on multitask learning in the stock market

Published: 12-10-2020| Version 1 | DOI: 10.17632/cw9tjsykxb.1
yi-jui chiu,
Heng-Chang Zhang,
Qing Wu


These historical transaction data of China's stock market are downloaded from the Choice Financial Data Terminal of Eastmoney, and which constitute the following stock index datasets and securities stock datasets, where the web address is ( The stock index datasets respectively include 500 historical trading days of Shanghai Securities Composite Index (SSEC), Shenzhen Composite Index (SZSC), Growth Enterprise Composite Index (CHINEXTC), and SSE SME Composite Index (SZSME). The time period is from September 1, 2014 to September 18, 2016. The historical data of every trading day contain 17 indicators: previous closing index, highest index, lowest index, closing index, index changing margin, index changing ratio, trading volume, trading amount, trading turnover rate, index amplitude, total market amount, circulating market amount, price-to-earnings ratio (P/E), price-to-book ratio (P/B), price-cash flow ratio (PCF), price-to-sales ratio (PS) and opening index. The securities stock datasets respectively include 400 historical trading days of Northeast Securities (DBZQ), Haitong Securities (HTZQ), Guangfa Securities (GFZQ), and Soochow Securities (DWZQ). The time period is from November 1, 2017 to June 24, 2019. The historical data of every trading day contain 12 indicators: previous closing price, highest price, lowest price, closing price, price changing margin, price changing ratio, trading turnover rate, trading volume, trading amount, total market amount, circulating market amount, and opening price.