Unlinked monotone regression

We consider so-called univariate unlinked (sometimes ``decoupled,'' or ``shuffled'') regression when the unknown regression curve is monotone. In standard monotone regression, one observes a pair where a response is linked to a covariate through the model , with the (unknown) monotone regression function and the unobserved error (assumed to be independent of ). In the unlinked regression setting one gets only to observe a vector of realizations from both the response and from the covariate where now . There is no (observed) pairing of and . Despite this, it is actually still possible to derive a consistent non-parametric estimator of under the assumption of monotonicity of and knowledge of the distribution of the noise . In this paper, we establish an upper bound on the rate of convergence of such an estimator under minimal assumption on the distribution of the covariate . We discuss extensions to the case in which the distribution of the noise is unknown. We develop a second order algorithm for its computation, and we demonstrate its use on synthetic data. Finally, we apply our method (in a fully data driven way, without knowledge of the error distribution) on longitudinal data from the US Consumer Expenditure Survey.
View on arXiv