The 23rd seminar series on Quantitative Methods and Data Analysis was successfully held

We are honored to invite Professor Tu Yundong from the Guanghua School of Management at Peking University. Professor Tu is one of the first group of "Rising Sun in the East" at Peking University's Guanghua School of Management and a young scholar of the Ministry of Education's "Yangtze River Scholars Award Program". He has published more than 30 papers in top international journals like the Journal of Econometrics, Econometric Reviews, and the Journal of Business and Economic Statistics.


图片.png


 The research topic presented by Professor Tu in this session is "Penetrating Sporadic Return Predictability". This work aims to predict returns by proposing a multi-structural change prediction regression that explored potential predictive factors. In the regression, the predictive factors exhibit low signal strength and varying degrees of persistence. To improve the prediction accuracy, breakpoint identification is performed before variable selection to achieve the effectiveness and heterogeneity of structural identification.


Professor Tu empirically tests the predictability of most variables on the US stock returns and finds that their predictability varies at different times. The  results show that the model can accurately identify actual breakpoints. In addition, in terms of out-of-sample prediction, the model also reveals varying degree of improvement compared to existing models.


During the seminar, Professor Tu answered and discussed questions raised by the participants ( (teachers and students) in detail. Finally, the event ended successfully with a lively discussion environment!