JOB MARKET PAPER
Stock Co-jump Networks with Mixed Membership
- with Guoli Liu, Xinghua Zheng and Yingying Li, 2024

Abstract: We build the stock co-jump network based on the high-frequency data to study the linkage between stocks. We propose a new model to involve the stock’s mixed membership structure and develop a statistical machine learning algorithm, Mixed-SCORE-DMP. We show that Mixed-SCORE-DMP is asymptotically consistent in estimating the mixed membership structures. Empirically, we find a statistically significant co-movement in the fundamentals of mixed membership firms. The purity of individual stocks has a strictly monotonic relationship with both volatility and the Sharpe ratio, and the peer momentum defined by the mixed membership has a stronger network momentum effect..

PUBLICATIONS

How to Dominate the Historical Average
- with Yingying Li, Kai Li and Jialin Yu, 2024, SSRN
- [Review of Financial Studies, forthcoming]

Abstract: We present a novel methodology for the out-of-sample forecast of the equity premium. Our approach matches the stability of the historical average while conservatively utilizing a predictor’s capacity to improve forecast accuracy. We demonstrate that, theoretically and empirically, our method dominates the historical average in forecast performance. Our methodology establishes a simple yet powerful paradigm for exploiting the real-time equity premium predictability derived from a predictor. Applications of our method reveal that many predictors can forecast the equity premium, and that parameter estimates in previous studies add value to out-of-sample forecasts.

WORKING IN PROGRESS

-Jumps, Overnight returns, Intraday Momentum, and Future Stock Returns
-Pay for Environment
- with Arthur Morris and Ruichao Zhu
-Non-financial Incentives, Production Efficiency, and Pricing Firm’s Green Actions\