Section 3.2.2 in the Portfolio Construction and Risk Management book gives a brief introduction to generative machine learning methods for investment market simulation, including variational autoencoders (VAEs), which can also be used to impute missing data.
This video presents the VAE examples of the Investment Simulation module, including multi-asset simulation as well as holiday, long-term, and alternatives missing data imputation.
The attached PDF provides a written summary of the perspectives and case studies.
See also this video presenting stationary transformations and the Fully Flexible Resampling method.
Access exclusive content
Paid subscribers have access to walkthroughs of exclusive case studies that include several advanced applications and Python code.
As a free subscriber, you can claim a one-time access to a paid post. Subscribe now or upgrade to a paid subscription to see the full video.
