On the 7th of December, the UK Campus had the pleasure of having Prof. Zhenya Liu to talk about his latest research project “A mispricing factor, IPCA and China A-Share Market”.
Prof. Liu is currently a full professor at the School of Finance, Renmin University of China, and CERGAM, Aix-Marseille University. He also held positions at the University of Birmingham, JP Morgan Futures (China) and the International Monetary Fund (IMF). He is a leading expert in financial econometrics and quantitative investments, and founded the graduate programs in Quantitative Investment at the Renmin University of China.
The presentation was attended by Peking University students and academics.
The seminar presented by Professor Liu was aimed to explain the issue surrounding returns’ predictability and empirical methodology in computing factor models. He started off the seminar by highlighting the importance of explaining stock returns with the right model; and emphasized the different approaches among researchers in finance and economics, practitioners, and statisticians/mathematicians. He pointed out that these three groups, despite the different background and way of approaching the same topic, should take factor analysis to the next level, by proposing a unified approach to explain financial markets.
Such unified approach was tackled by his research project, where he proposed an instrumented principal component analysis (IPCA) model to better capture asset pricing. Asset mispricing, indeed, is very pronounced in the Chinese A-share market, dominated by existence of individual investors–retail investors hold 99.74% stock accounts– and trade frictions.
Next, Professor Liu presented and discussed exhaustively the proposed methodology that estimates the linearly parameterized mapping together with the latent factors by least squares similar to principal component analysis (PCA). Such methodology is more flexible due to its time-variability and increased economic content from instrumental information and gives economic interpretability to the estimates of latent factors. Professor Liu also guided the audience to the construction of the mispricing factor, a crucial component in his proposed model.
Finally, Professor Liu provided the empirical results of his methodology, and compared it with current popular factor models. Overall, his model outperforms the current popular factor models in explaining China’s A-shares market due to the irrational behavior of investors. His tangency portfolio achieves an out-of-sample of 2.42.
Such interesting and engaging seminar left a lasting impression on participants. The audience posed questions during the session, while Professor Liu shared episodes from his experiences as hedge fund manager, and highlighted the importance of developing strong quantitative skills to succeed in such an industry.
We thank Professor Liu for his wonderful and illuminating seminar, and for sharing his knowledge and expertise with us. We definitely learned a lot about returns’ predictability, and we hope to welcome him back to our UK Campus soon.
