A joint estimation approach for monotonic regression functions in general dimensions

Rohrbeck, Christian and Costain, Deborah (2025) A joint estimation approach for monotonic regression functions in general dimensions. Scandinavian Journal of Statistics. ISSN 0303-6898

Full text not available from this repository.

Abstract

Regression analysis under the assumption of monotonicity is a well-studied statistical problem and has been used in a wide range of applications. However, there remains a lack of a broadly applicable methodology that permits informa-tion borrowing, for efficiency gains, when jointly estimating multiple monotonic regression functions. We fill this gap in the literature and introduce a methodology which can be applied to both fixed and random designs and any number of explanatory variables (regressors). Our framework penalizes pairwise differences in the values of the monotonic function estimates, with the weight of penalty being determined, for instance, based on a statistical test for equivalence of functions at a point. Function estimates are subsequently derived using an iterative optimization routine which updates the individual function estimates in turn until convergence.Simulation studies for normally and binomially distributed response data illustrate that function estimates are improved when similarities between functions exist, and are not over-smoothed otherwise. We further apply our methodology to analyze two public health data sets: neonatal mortality data for Porto Alegre, Brazil, and stroke patient data for North West England.

Item Type:
Journal Article
Journal or Publication Title:
Scandinavian Journal of Statistics
Uncontrolled Keywords:
Research Output Funding/no_not_funded
Subjects:
?? convex optimization, likelihood ratio test, monotonic regression,public health, shape constraintsno - not fundedstatistics and probabilitystatistics, probability and uncertainty ??
ID Code:
228514
Deposited By:
Deposited On:
27 Mar 2025 09:55
Refereed?:
Yes
Published?:
Published
Last Modified:
27 Mar 2025 09:55