Stanford University uses deep neural networks to estimate asset pricing for individual stock returns, taking advantage of a vast amount of conditioning information while keeping a fully flexible form and accounting for time variations. Their key innovations include using the fundamental no-arbitrage condition as the criterion function, constructing the most informative test assets with an adversarial approach, and extracting the states of the economy from many macroeconomic time series.

• How Stanford University’s asset pricing model outperforms all out-of-sample benchmark approaches in terms of Sharpe ratio, explained variations, and pricing errors

• How to identify the key factors that drive asset prices
  
  
• The asset pricing model that yields profitable machine learning investment strategies

Content

content goes here