Diagnosing financial anomalies in China: liquidity risk, information efficiency, and macroeconomic linkages

Abstract

The pricing of financial assets with respect to their fundamental values, as well as the presumption of the absence of return predictability given market efficiency have long been basic tenets of financial economics. Since the 1960s, measuring equity premiums with respect to their risk exposures have been a recognized common practice in return prediction. Anomalies, however, often occur that challenge the central predictions of efficient market hypothesis and asset pricing theories. This thesis examines the predictability of equity returns in China and explores three fundamental questions: Does anomalies with respect to return predictability exist, what are the important predictive variables, and what are the rationale behind the variables' predictive ability? By relying on both conventional econometric models and cutting-edge machine-learning techniques, I seek to determine the driving factors for return predictability and pricing anomalies in the Chinese market. Specifically, I design three empirical studies which propose that cross-sectional differences in average stock returns are determined not only by the market risk and firm-level risk exposures, as prescribed by conventional multi-factor asset pricing models, but also by other risk factors such as the liquidity, information efficiency and cyclic macroeconomic conditions. By formulating analytical methodologies that considers fundamental, technical and regime predictors, and assessing their predictive ability over heterogeneous horizons, this thesis thus complements the literature by providing new empirical evidence on how the distinctive Chinese market structure and investor reaction to new information impact the return predictability and market efficiency in this quickly emerging market.

Date of Award	Jul 2024
Original language	English
Awarding Institution	University of Nottingham
Supervisor	Chew Chua (Supervisor) & Arijit Mukherjee (Supervisor)