Randomness in Finance is traditionally studied in stochastic terms, though there exists, since long, another theoretical framework that also adresses the notion of random series: the Algorithmic Complexity Theory. Following this latter framework, many applications have been developed in biology, in computer science and other scientific fields, while we are still at the beginning of the embedness of algorithmic theory in finance.

Machine learning technics are introduced in finance since years. 

Classification techniques are generally used in corporate default predicting, in fraud detecting, in credit scoring, and asset pricing problems.

We try to use machine learning technics to explain RSE scoring methodologies and explain some RSE empirical observations.

I’m also interested by the impact of derivative products on spot market volatilily. This is a well debated topics in finance, many available theoretical and empirical studies. However, their results often diverge from model to model and from database to database. With the development of high frequency data uses in finance, can we add new contributions to this debate?