Monotone least squares and isotonic quantiles

DOI10.1214/19-EJS1659zbMATH Open1434.62056arXiv1901.02398MaRDI QIDQ2283585FDOQ2283585

Authors: Alexandre Mösching, Lutz Dümbgen 

Publication date: 3 January 2020

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

Abstract: We consider bivariate observations $(X_1,Y_1),ldots,(X_n,Y_n)$ such that, conditional on the $X_i$, the $Y_i$ are independent random variables with distribution functions $F_{X_i}$, where $(F_x{)}_x$ is an unknown family of distribution functions. Under the sole assumption that $xmapsto{F}_x$ is isotonic with respect to stochastic order, one can estimate $(F_x{)}_x$ in two ways: (i) For any fixed $y$ one estimates the antitonic function $xmapsto{F}_x(y)$ via nonparametric monotone least squares, replacing the responses $Y_i$ with the indicators $1_{[Y_{i}ley]}$. (ii) For any fixed $$ one estimates the isotonic quantile function $$ via a nonparametric version of regression quantiles. We show that these two approaches are closely related, with (i) being more flexible than (ii). Then, under mild regularity conditions, we establish rates of convergence for the resulting estimators $hat{F}_x(y)$ and $$, uniformly over $(x,y)$ and $$ in certain rectangles as well as uniformly in $y$ or $$ for a fixed $x$.

Full work available at URL: https://arxiv.org/abs/1901.02398

Recommendations

• An improved monotone conditional quantile estimator
• Distribution-free properties of isotonic regression
• Convergence rates of monotone conditional quantile estimators
• Order-Preserving Nonparametric Regression, With Applications to Conditional Distribution and Quantile Function Estimation
• Least-squares estimation of two-ordered monotone regression curves

zbMATH Keywords

regression quantilesuniform consistencystochastic order

Mathematics Subject Classification ID

Nonparametric regression and quantile regression (62G08) Measures of association (correlation, canonical correlation, etc.) (62H20) Inequalities; stochastic orderings (60E15)

Cites Work

• Regression Quantiles
• Title not available (Why is that?)
• About monotone regression quantiles.
• Extensions of smoothing via taut strings
• Inferences Under a Stochastic Ordering Constraint
• New goodness-of-fit tests and their application to nonparametric confidence sets
• On Models and Hypotheses with Restricted Alternatives
• Maximum Likelihood Estimation of the Distributions of Two Stochastically Ordered Random Variables
• Monotone percentile regression
• An improved monotone conditional quantile estimator
• Algorithms in order restricted statistical inference and the Cauchy mean value property
• A uniform bound for the tail probability of Kolmogorov-Smirnov statistics
• The asymptotic behavior of monotone percentile regression estimates
• Characterizing the optimal solutions to the isotonic regression problem for identifiable functionals
• Title not available (Why is that?)

Cited In (18)

• Uniform estimation of isobars
• Monotone support vector quantile regression
• Regression diagnostics meets forecast evaluation: conditional calibration, reliability diagrams, and coefficient of determination
• Least-squares estimation of two-ordered monotone regression curves
• Easy Uncertainty Quantification (EasyUQ): Generating Predictive Distributions from Single-Valued Model Output
• Estimation and bootstrap for stochastically monotone Markov processes
• Distribution‐free Estimation of a Monotonic pth Quantile Function
• New challenges in the interplay between finance and insurance. Abstracts from the workshop held October 1--6, 2023
• Consistent Estimation of Distribution Functions under Increasing Concave and Convex Stochastic Ordering
• Forecast evaluation of quantiles, prediction intervals, and other set-valued functionals
• Isotonic regression for elicitable functionals and their Bayes risk
• Order-Preserving Nonparametric Regression, With Applications to Conditional Distribution and Quantile Function Estimation
• Accelerating the pool-adjacent-violators algorithm for isotonic distributional regression
• Convergence rates of monotone conditional quantile estimators
• Characterizing the optimal solutions to the isotonic regression problem for identifiable functionals
• Distributional (Single) Index Models
• Estimation and convergence rates in the distributional single index model
• Isotonic quantile regression: asymptotics and bootstrap

This page was built for publication: Monotone least squares and isotonic quantiles

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2283585)