Rongzhao Ou, PhD, CFA, FRM

Rongzhao Ou, PhD, CFA, FRM

Publications

You can also find my articles on my Google Scholar profile.

Are analysts’ Forecasts Reliable? A Machine Learning-Based Analysis of the Target Price Accuracy

Published in Journal of Behavioral Finance, 2024

Abstract: This paper examines the accuracy of target price forecasts made by sell-side analysts, focusing on predicting target price accuracy using machine learning approaches. Utilizing a dataset of target price forecasts for U.S. listed companies from 1999 to 2021, we employ ensemble methods and incorporate market-level, firm-level, and analyst-level information to predict target price accuracy in terms of target price errors and target price achievement. The long-short portfolio constructed based on our predictions significantly outperform the benchmark in terms of cumulative return and Sharpe ratio.

Recommended citation: Ou, Rongzhao & Wang, Qiao. (2024). "Are analysts’ Forecasts Reliable? A Machine Learning-Based Analysis of the Target Price Accuracy." Journal of Behavioral Finance. 1–17.

Price Discovery or Overreaction? A Study on the Reaction of Asia Pacific Country ETFs to the US Stock Market

Published in Investment Analysts Journal, 2023

Abstract: Despite the presence of arbitrage mechanisms, large premiums (or discounts) for Asia Pacific country ETFs in the US market could still exist in the short run due to the time gap between trading hours of the US and Asia Pacific markets. The price of a country ETF is not solely determined by net asset value but is also affected by information released during US trading hours. In this study, I examine six Asia Pacific country ETFs from 2006 to 2020, using linear regression as well as tree-based ensemble methods to predict the next-day return of the net asset value by analysing information from country ETFs and the S&P 500 Index. The results indicate that the trading hours of local markets significantly influence the predictive power of country ETFs and the S&P 500 Index. The findings suggest that the returns of these ETFs do not necessarily overreact to the US market but instead reflect short-term expectations of the performance of underlying indices. Furthermore, I extend the model to analyse overnight and daytime returns and identify the price correction that occurs during daytime trading hours.

Recommended citation: Ou, Rongzhao. (2023). "Price Discovery or Overreaction? A Study on the Reaction of Asia Pacific Country ETFs to the US Stock Market." Investment Analysts Journal. 52:4, 349-364.

Efficiency of Different VaR Computing Methods–Based on the Performance of VaR in Upward, Downward and Swinging Trends of Stock Market

Published in 7nd China Lixin Risk Management Conference, 2014

Abstract: Plenty of researches have shown that the efficiency of Value-at-Risk in risk management may change as the calculation methods and the market condition change. In this paper, MA-60 and MA-250 were used to separate upward, downward and swinging trends. Historical simulation, delta-normal method and Monte Carlo simulation are used to calculate VaR, respectively. Kupiec test and Christoffersen test are used to backtest the results. In delta-normal method and Monte Carlo simulation, EQMA, EWMA and GARCH are used to estimate volatility. Empirical results indicate that VaR with historical simulation tends to overestimate risk. The sharp peak and heavy tail characteristics of returns result in that delta-normal method underperforms Monte Carlo Simulation. VaR calculated by Monte Carlo simulation with volatility estimated by GARCH performs the best in Kupeic test and Christoffersen test.

Recommended citation: Ou, Rongzhao. (2014). "Efficiency of Different VaR Computing Methods–Based on the Performance of VaR in Upward, Downward and Swinging Trends of Stock Market." 7nd China Lixin Risk Management Conference. 2014, 261-270.

A Study on the Relationship among the Leading Actors, Directors, and the Box Office Income of a Film – Based on Multiple Linear Regression Model

Published in Information Management, Innovation Management and Industrial Engineering (ICIII), 2013

Abstract: This paper makes a comprehensive and systematic study of the impact of the relationship among the leading actor, director, and the box office income of a film. We give scores according to the awareness and attention of the leading actor and director from a very influential database, and make multiple linear regression using these data. The result shows that the box office income is related to the expertise of the director. The better the director is, the higher the box office income will be. However, the box office income has little relationship with the leading actor. It’s an interesting result because usually many persons believe the quality of actors is the biggest factor for the revenue of the film box office. Therefore, to pursuit higher box office income and better awareness, movie companies’ administrators should consider on hiring famous directors in the first place.

Recommended citation: Yang, Yongbin & Ou, Rongzhao. (2013) "A Study on the Relationship among the Leading Actors, Directors, and the Box Office Income of a Film – Based on Multiple Linear Regression Model." Information Management, Innovation Management and Industrial Engineering (ICIII) 6(1), 469-471.