How fast is the bandit?

Damien Lamberton (1) , Gilles Pagès (2)

1 LAMA - Laboratoire d'Analyse et de Mathématiques Appliquées
2 LPMA - Laboratoire de Probabilités et Modèles Aléatoires

In this paper we investigate the rate of convergence of the so-called two-armed bandit algorithm in a financial context of asset allocation. The behaviour of the algorithm turns out to be highly non-standard: no CLT whatever the time scale, possible existence of two rate regimes.

Mots clés: asset allocation; Two-armed bandit algorithm; Stochastic Approximation; learning automata; asset allocation.

Domaines: Probabilités [math.PR]

Damien Lamberton, Gilles Pagès. How fast is the bandit?. Stochastic Analysis and Applications, 2008, 26 (3), pp.603-623. ⟨10.1080/07362990802007202⟩. ⟨hal-00012182⟩