Dependence of variance on covariate design in nonparametric link regression
arXiv:2012.13106
Abstract
This paper discusses a design-dependent nature of variance in nonparametric link regression aiming at predicting a mean outcome at a link, i.e., a pair of nodes, based on currently observed data comprising covariates at nodes and outcomes at links.
24 pages, 11 figures, accepted to Statistics and Probability Letters