Facing the challenges of large-dimensional characteristics and the limitations of ad-hoc sparsity assumption, this paper constructs a portfolio of systematic risk factors with stronger asset pricing capability and broader representativeness for the A-share market. Drawing on the implicit factor structure with reference to the cutting-edge research in the field of large-dimensional factor statistical inference, this study conducts a comprehensive deconstruction and analysis of the corresponding factor characteristics and their associated risk compensation structure. The results indicate that: 1) Compared with the CH-〖KG*4〗4 and Fama-French multi-factor pricing models, the large-dimensional high-frequency RP-PCA factor portfolio captures more systematic risk characteristics and exhibits better pricing performance along with a more robust time-varying loading structure; 2) The A-share market contains five stable systematic risk factors. Apart from the market factor, the other four can be approximated by the portfolios of six specific sub-sectors, predominantly represented by the finance sector; 3) Systematic risk in the A-share market is not fully priced, and there is significant asymmetry in risk exposure. Overall, investors tend to be more concerned about risks when the market rises while individual stocks fall, as well as during simultaneous declines in both the market and individual stocks.