Financial factors selection with knockoffs: fund replication, explanatory and prediction networks

Damien Challet (1) , Christian Bongiorno (1) , Guillaume Pelletier (2)

1 MICS - Mathématiques et Informatique pour la Complexité et les Systèmes
2 BNP-Paribas

We apply the knockoff procedure to factor selection in finance. By building fake but realistic factors, this procedure makes it possible to control the fraction of false discovery in a given set of factors. To show its versatility, we apply it to fund replication and to the inference of explanatory and prediction networks.

Domains: Applications [stat.AP]; Human genetics; Data Analysis, Statistics and Probability [physics.data-an]; Computer science; Computational Finance [q-fin.CP]

Damien Challet, Christian Bongiorno, Guillaume Pelletier. Financial factors selection with knockoffs: fund replication, explanatory and prediction networks. 2021. ⟨hal-03165842⟩