Random Forest based Qantile Oriented Sensitivity Analysis indices estimation

Kévin Elie-Dit-Cosaque (1, 2, 3) , Véronique Maume-Deschamps (1, 2)

1 ICJ - Institut Camille Jordan
2 PSPM - Probabilités, statistique, physique mathématique
3 SCOR SE [Paris]

We propose a random forest based estimation procedure for Quantile-Oriented Sensitivity Analysis—QOSA. In order to be efficient, a cross-validation step on the leaf size of trees is required. Our full estimation procedure is tested on both simulated data and a real dataset. Our estimators use either the bootstrap samples or the original sample in the estimation. Also, they are either based on a quantile plug-in procedure (the R-estimators) or on a direct minimization (the Q-estimators). This leads to 8 different estimators which are compared on simulations. From these simulations, it seems that the estimation method based on a direct minimization is better than the one plugging the quantile. This is a significant result because the method with direct minimization requires only one sample and could therefore be preferred.

Mots clés: Quantile Oriented Sensitivity Analysis; Random forest; Cross validation; Out of Bag samples

Domaines: Probabilités [math.PR]; Statistiques [math.ST]

Kévin Elie-Dit-Cosaque, Véronique Maume-Deschamps. Random Forest based Qantile Oriented Sensitivity Analysis indices estimation. Computational Statistics, 2024, ⟨10.1007/s00180-023-01450-5⟩. ⟨hal-03151021v2⟩