Significance testing in non-sparse high-dimensional linear models
Significance testing in non-sparse high-dimensional linear models is a scholarly work, published in 2018 in ''Electronic Journal of Statistics''. The main subjects of the publication include dependent and independent variables, regularization, heteroscedasticity, type I and type II errors, missing data, causal inference, mathematics, inference, and algorithm. The authors show that existing inferential methods are sensitive to the sparsity\nassumption, and may, in turn, result in the severe lack of control of Type-I\nerror.