The single-index model is a statistical model for intrinsic regression where responses are assumed to depend on a single yet unknown linear combination of the predictors, allowing to express the regression function as E[Y vertical bar X] = f (< v,X >) for some unknown index vector v and link function f. Conditional methods provide a simple and effective approach to estimate v by averaging moments of X conditioned on Y, but depend on parameters whose optimal choice is unknown and do not provide generalization bounds on f. In this paper we propose a new conditional method converging at root n rate under an explicit parameter characterization. Moreover, we prove that polynomial partitioning estimates achieve the 1-dimensional min-max rate for regression of Holder functions when combined to any root n-convergent index estimator. Overall this yields an estimator for dimension reduction and regression of single-index models that attains statistical optimality in quasilinear time.
Conditional regression for single-index models / A. Lanteri, M. Maggioni, S. Vigogna. - In: BERNOULLI. - ISSN 1350-7265. - 28:4(2022 Nov), pp. 3051-3078. [10.3150/22-BEJ1482]
Conditional regression for single-index models
A. Lanteri
Primo
;
2022
Abstract
The single-index model is a statistical model for intrinsic regression where responses are assumed to depend on a single yet unknown linear combination of the predictors, allowing to express the regression function as E[Y vertical bar X] = f (< v,X >) for some unknown index vector v and link function f. Conditional methods provide a simple and effective approach to estimate v by averaging moments of X conditioned on Y, but depend on parameters whose optimal choice is unknown and do not provide generalization bounds on f. In this paper we propose a new conditional method converging at root n rate under an explicit parameter characterization. Moreover, we prove that polynomial partitioning estimates achieve the 1-dimensional min-max rate for regression of Holder functions when combined to any root n-convergent index estimator. Overall this yields an estimator for dimension reduction and regression of single-index models that attains statistical optimality in quasilinear time.File | Dimensione | Formato | |
---|---|---|---|
BEJ1482.pdf
accesso aperto
Tipologia:
Publisher's version/PDF
Dimensione
842.61 kB
Formato
Adobe PDF
|
842.61 kB | Adobe PDF | Visualizza/Apri |
2002.10008.pdf
accesso aperto
Tipologia:
Pre-print (manoscritto inviato all'editore)
Dimensione
939.65 kB
Formato
Adobe PDF
|
939.65 kB | Adobe PDF | Visualizza/Apri |
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.