The Chinese stock market is replete with numerous omitted variables that can introduce biases in the standard estimation of risk premiums when traditional linear asset pricing models are applied. The three-pass method enables the estimation of risk premiums for observable factors even when not all relevant factors are explicitly specified or observed within the model. Accordingly, we have applied this method to construct portfolios with stocks from China’s A-share market as the test assets. Empirical research findings indicate that the three-pass method could be more effective than traditional linear asset pricing models in estimating risk premiums.
Understanding the risk-return trade-off in the options market is difficult with standard observable factor models. We examine whether dynamic factor loadings improve the explanation of the cross-section of delta-hedged option returns. We apply instrumented principal component analysis (IPCA) to major exchange-traded option products in China. IPCA allows conditional betas to vary with option characteristics, which addresses short contract lifespans and rapidly evolving characteristics. We find that a three-factor IPCA model explains 19.30% of variation in individual option returns and 99.23% in managed portfolios. A bootstrap asset-pricing test fails to reject a zero-alpha restriction. Observable factor models with characteristic-instrumented dynamic betas achieve performance close to IPCA, highlighting the importance of time-varying loadings. We also provide subsample and characteristic-importance analyses.
This paper applies an integrated Bayesian Model Averaging (BMA) framework across 2^26 specifications to examine asset-pricing uncertainty in China's A-share market. We document four main findings. First, model uncertainty in China is substantially higher than in mature markets. Second, posterior evidence strongly favors time-varying specifications. Third, the integrated BMA model delivers superior Sharpe ratios and downside risk performance relative to leading benchmarks. Fourth, decomposing model uncertainty along three dimensions reveals that: (i) factor risk is materially elevated, with integrated predictive variance far exceeding historical sample variance out of sample; (ii) covariance-entropy dynamics show that model disagreement intensifies sharply during periods of market stress; and (iii) disagreement over mispricing, factor loadings, and risk premia is concentrated in interactive and time-varying components of expected returns. Overall, model uncertainty plays a central role in shaping risk assessment, factor interpretation, and portfolio decisions in China's equity market.