Machine learning methods for identifing investment factors
There are now hundreds of different signals in the literature for predicting the return of a stock. These have been tested extensively, especially in regression models. The goal of this master thesis is to present a:
- comparative analysis of different machine learning methods for the problem of empirical valuation of stock prices.
- In doing so, an extensive database is used to show that these methods outperform leading regression-based strategies from the literature in some cases.
- Furthermore, the best method will be identified and it will be explained how the high predictive power is achieved.
- Furthermore, it will be shown which variables are most important for the prediction.
References:
Gu, Shihao; Kelly, Bryan; Xiu, Dacheng (2020): Empirical asset pricing via machine learning, in: The Review of Financial Studies, Vol. 33.5, S. 2223-2273.