Threshold regression with nonparametric sample splitting

This paper develops a threshold regression model where an unknown relationship between two variables nonparametrically determines the threshold. We allow observations to be cross-sectionally dependent so that can applied determine spatial border for sample splitting over random field. derive uniform rate of convergence and nonstandard limiting distribution nonparametric estimator. also obtain root-n consistency asymptotic normality coefficient illustrate empirical relevance this new by estimating tipping point in social segregation problems as function demographic characteristics; determining metropolitan area boundaries using nighttime light intensity collected from satellite imagery.