This paper consider penalized empirical loss minimization of convex loss functions with unknown non-linear target functions. Using the elastic net penalty we establish a finite sample oracle inequality which bounds the loss of our estimator from above with high probability. If the unknown target...

We propose a new estimator, the thresholded scaled Lasso, in high dimensional threshold regressions. First, we establish an upper bound on the sup-norm estimation error of the scaled Lasso estimator of Lee et al. (2012). This is a non-trivial task as the literature on highdimensional models has...

While variable selection and oracle inequalities for the estimation and prediction error have received considerable attention in the literature on high-dimensional models, very little work has been done in the area of testing and construction of confidence bands in high-dimensional models....

The purpose of this paper is to develop and estimate a new equilibrium model of public housing that acknowledges the fact that the demand for public housing may exceed the available supply. We show that ignoring these supply side restrictions leads to an inconsistent estimator of household...

We develop a new dynamic equilibrium model with heterogeneous households that captures the most important frictions that arise in housing rental markets and explains the political popularity of affordable housing policies. We estimate the model using data collected by the New York Housing...

Many urban school districts in the United States and OECD countries confront the necessity of closing schools due to declining enrollments. To address this important policy question, we formulate a sequential game where a superintendent is tasked with closing down a certain percentage of student...