Finite sample behavior of a sieve profile estimator in the single index mode

DOI10.1214/15-EJS1079zbMATH Open1326.62046OpenAlexW2176178582MaRDI QIDQ895013FDOQ895013

Authors Andreas Andresen 

Publication date 25 November 2015

Published in Electronic Journal of Statistics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1406.4052

zbMATH Keywords

sievealternating minimizationprofile estimatorsingle indexalternating maximizationprojection pursuit procedure

Mathematics Subject Classification ID

Point estimation (62F10) Parametric tolerance and confidence regions (62F25) Estimation in multivariate analysis (62H12) Generalized linear models (logistic models) (62J12)

Abstract: We apply the results of Andresen A. and Spokoiny V. on profile M-estimators and the alternating maximization procedure to analyse a sieve profile quasi maximum likelihood estimator in the single index model with linear index function. The link function is approximated with (C^3)-Daubechies-wavelets with compact support. We derive results like Wilks phenomenon and Fisher Theorem in a finite sample setup. Further we show that an alternation maximization procedure converges to the global maximizer and assess the performance of a projection pursuit procedure in that context. The approach is based on showing that the conditions of Andresen A. and Spokoiny V. on profile M-estimators and the alternating maximization procedure can be satisfied under a set of mild regularity and moment conditions on the index function, the regressors and the additive noise. This allows to construct nonasymptotic confidence sets and to derive asymptotic bounds for the estimator as corollaries.

Recommendations

• Sieve maximum likelihood estimator for functional probit model
• Series estimation for single-index models under constraints
• Critical dimension in profile semiparametric estimation
• M-estimators for single-index model using B-spline
• Adaptive estimation in the single-index model via oracle approach

Cites work

• ASYMPTOTIC DISTRIBUTIONS FOR TWO ESTIMATORS OF THE SINGLE-INDEX MODEL
• An Adaptive Estimation of Dimension Reduction Space
• Convergence of an alternating maximization procedure
• Convergence rates and asymptotic normality for series estimators
• Critical dimension in profile semiparametric estimation
• Direct estimation of the index coefficient in a single-index model
• Estimating the direction in which a data set is most interesting
• Introduction to empirical processes and semiparametric inference
• Learning without concentration
• Majorizing measures: The generic chaining
• On a conjecture of Huber concerning the convergence of projection pursuit regression
• On methods of sieves and penalization
• Optimal rates of convergence for nonparametric estimators
• Optimal smoothing in single-index models
• Parametric estimation. Finite sample theory
• Projection pursuit
• Semiparametric Estimation of Index Coefficients
• Semiparametric least squares (SLS) and weighted SLS estimation of single-index models
• The Macroeconomy as a Network of Money-Flow Transfer Functions
• The sizes of compact subsets of Hilbert space and continuity of Gaussian processes
• User-friendly tail bounds for sums of random matrices
• Wavelets on the interval and fast wavelet transforms

Cited in

(2)

• Critical dimension in profile semiparametric estimation
• Alternating least squares in generalized linear models

This page was built for publication: Finite sample behavior of a sieve profile estimator in the single index mode

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q895013)