Congratulations to Oksana Bashchenko for the outstanding and successful defense of her PhD thesis on "Three applications of machine learning techniques to empirical asset pricing."In her first paper, she develops a new method that detects jumps nonparametrically in financial time series and significantly outperforms the current benchmark on simulated data. In the second paper, she builds on this methodology for detecting asset bubbles using a neural network. In the third paper, she proposes a new methodology to construct interpretable, fundamental-based pricing factors from news to explain Bitcoin returns.The first two papers rely on the long short-term memory (LSTM) neural network, trained on labelled simulated data. In the third paper, sentiment factors are identified using a natural language processing algorithm (SBERT network).On the picture, from left to right, me (thesis supervisor), Oksana, Christian Zehnder, and Roxana Mihet, both members of the thesis committee (with Pierre Collin-Dufresne).#HECLausanne