Lunyi Wang

We study how investor composition shapes the market effects of corporate information manipulation. A firm can add a noisy distortion with a positive mean to a public signal. Sophisticated investors correctly understand the manipulated reporting process, whereas unsophisticated investors treat the report as unbiased. A larger sophisticated share attenuates the price response to the report, reduces the expected price effect of manipulation, and eventually deters manipulation. Conditional on manipulation, heterogeneous interpretations allocate the risky asset more heavily to unsophisticated investors after favorable reports and generate an objective risk-allocation loss that is hump-shaped in the sophisticated share.

This paper studies how probability distortion affects the sensitivity of asset prices to order flow. We develop a simple asset-pricing model in which optimistic investors overweight favorable outcomes when evaluating risky assets. This distortion generates a flat segment in their demand, implying that equilibrium prices can become locally insensitive to liquidity shocks. As a result, the model produces price inertia: over a range of order imbalances, prices do not respond to order flow. The width of this inelastic region increases with investor optimism and the fraction of optimistic traders. When investors update their beliefs dynamically in response to realized outcomes, the model endogenously generates regimes of low and high volatility. The analysis provides a behavioral explanation for price inertia, sudden market adjustments, and volatility clustering.

In this paper, we examine the conditions under which the capital asset pricing model (CAPM) holds with heterogeneous probability weighting. Using the generalised Wang transform within rank‐dependent expected utility, we show that CAPM holds for heterogeneous risk‐averse investors, while the security market line theorem (SMLT) applies to heterogeneous loss‐averse investors. However, CAPM under loss aversion requires homogeneous investors. Revisiting skewness pricing, we find that probability weighting, rather than the S‐shaped value function, drives skewness overpricing. The preference for skewed assets stems from the high distorted mean under probability weighting.