VECTOR QUANTILE REGRESSION: AN OPTIMAL TRANSPORT APPROACH

• 成果类型: Article
• 署名作者: Carlier, Guillaume; Chernozhukov, Victor; Galichon, Alfred
• 署名单位: Universite PSL; Universite Paris-Dauphine; Massachusetts Institute of Technology (MIT); Massachusetts Institute of Technology (MIT); New York University; New York University
• 刊物名称: ANNALS OF STATISTICS
• ISSN/ISSBN: 0090-5364
• DOI: 10.1214/15-AOS1401
• 发表日期: 2016
• 页码: 1165-1192
• 关键词: multivariate quantiles

摘要

We propose a notion of conditional vector quantile function and a vector quantile regression. A conditional vector quantile function (CVQF) of a random vector Y, taking values in R-d given covariates Z = z, taking values in R-k, is a map u bar right arrow Q(Y vertical bar Z) (u, z), which is monotone, in the sense of being a gradient of a convex function, and such that given that vector U follows a reference non-atomic distribution F-U, for instance uniform distribution on a unit cube in Rd, the random vector Q(Y vertical bar Z) (U, z) has the distribution of Y conditional on Z = z. Moreover, we have a strong representation, Y = Q(Y vertical bar Z) (U, Z) almost surely, for some version of U. The vector quantile regression (VQR) is a linear model for CVQF of Y given Z. Under correct specification, the notion produces strong representation, Y = beta(U)(inverted perpendicular) f (Z), for f (Z) denoting a known set of transformations of Z, where u bar right arrow beta(u)(inverted perpendicular) f (Z) is a monotone map, the gradient of a convex function and the quantile regression coefficients u bar right arrow beta(u) have the interpretations analogous to that of the standard scalar quantile regression. As f (Z) becomes a richer class of transformations of Z, the model becomes nonparametric, as in series modelling. A key property of VQR is the embedding of the classical Monge Kantorovich's optimal transportation problem at its core as a special case. In the classical case, where Y is scalar, VQR reduces to a version of the classical QR, and CVQF reduces to the scalar conditional quantile function. An application to multiple Engel curve estimation is considered.