Search

This paper proposes a cross‐section long short‐term memory (CS‐LSTM) factor model to explore the possibility of estimating expected returns in the Chinese stock market. In contrast to previous machine‐learning‐based asset pricing models that make predictions directly on equity returns, CS‐LSTM estimates are based on predictions of slope terms from Fama–MacBeth cross‐section regressions using 16 stock characteristics as factor loadings. In line with previous studies in the context of the Chinese market, we find illiquidity and short‐term momentum to be the most important factors in describing asset returns. By using 274 value‐weighted portfolios as test assets, we systematically compare the performances of CS‐LSTM and three other candidate models. Our CS‐LSTM model consistently delivers better performance than the candidate models and beats the market at all different levels of transaction costs. In addition, we observe that assets with smaller cap are favored by the model. By repeating the empirical analysis based on the top 70% of stocks, our CS‐LSTM model remains robust and consistently provides significant market‐beating performance. Our findings from the CS‐LSTM model have practical implications for the future development of the Chinese stock market and other emerging markets. [ABSTRACT FROM AUTHOR]

Detecting pipe bursts in water distribution systems (WDSs) is of critical importance for urban infrastructure maintenance. A pipe burst can be detected from measurements that are continuously collected from hydraulic meters installed in WDSs, with widely accepted statistical process control techniques. However, the significant autocorrelation inevitably embedded in the continuously collected hydraulic measurements makes it extremely difficult for existing methods to accurately estimate the breakout time and the magnitude of a burst. To overcome the limitation, this paper proposes a new method to model the autocorrelation patterns with functional basis expansion. Functional regression is adopted to detect the pipe burst by decomposing the hydraulic measurements into three components: the normal components, the burst-induced anomaly component, and noises. A regularized estimation algorithm is developed to identify the three components by incorporating the knowledge of the impacts of bursts on the autocorrelation patterns in hydraulic measurements. A simulated water distribution network is built through EPANET. Analysis results based on the simulated data show that the proposed method not only outperforms existing methods with higher burst detectability and lower false alarm rate, but can also estimate the burst starting time, and magnitude estimation. [ABSTRACT FROM AUTHOR]