Nonparametric Regression Based on Ranks
摘要
Consider the linear regression model \(\displaystyle Y_i=\beta _0+{\mathbf x}_i^{\top }{\boldsymbol {\beta }}+e_i,\quad i=1,\ldots ,N \) where \(\beta _0\in {\mathbb R}_1, \ \boldsymbol {\beta }\in {\mathbb R}_p\) are unknown parameters and e 1, …, e N are independent errors, identically distributed according to a continuous d.f. F and \({\mathbf x}_i\in {\mathbb R}_p\) are given regressors, i = 1, …, N.