Evaluation for moments of a ratio with application to regression estimation

Paul Doukhan (1, 2) , Gabriel Lang (3)

1 SAMOS - Statistique Appliquée et MOdélisation Stochastique
2 CES - Centre d'économie de la Sorbonne
3 MIA-Paris - Mathématiques et Informatique Appliquées

Many problems involve ratios in probability or in statistical applications. We aim at approximating the moments of such ratios under specific assumptions. Using ideas from Collomb (1977) we propose sharper bounds for the moments of randomly weighted sums which also may appear as a ratio of two random variables. Suitable applications are given in more detail here in the fields of functional estimation, in finance and for censored data analysis. Several weak dependence dependence situations are considered.

Domaines: Statistiques [math.ST]; Théorie [stat.TH]

Paul Doukhan, Gabriel Lang. Evaluation for moments of a ratio with application to regression estimation. 2008. ⟨hal-00276881⟩