Regression rank scores in nonlinear models
Regression rank scores in nonlinear models
<!-- *** Custom HTML *** --> Consider the nonlinear regression model <i>Y</i><sub><i>i</i></sub>=<i>g</i>(<b>x</b><sub><i>i</i></sub>, <b><i>θ</i></b>)+<i>e</i><sub><i>i</i></sub>, <i>i</i>=1, …, <i>n</i> with <b>x</b><sub><i>i</i></sub>∈ℝ<sup><i>k</i></sup>, <b><i>θ</i>=(<i>θ</i></b><sub><b>0</b></sub><b>, <i>θ</i></b><sub><b>1</b></sub><b>, …, <i>θ</i></b><sub><b><i>p</i></b></sub><b>)</b><sup><b><i>′</i></b></sup><b>∈Θ</b> (compact in ℝ<sup><i>p</i>+1</sup>), where <i>g</i>(<b>x</b>, <b><i>θ</i></b>)=<i>θ</i><sub>0</sub>+<i>g̃</i>(<b>x</b>, <i>θ</i><sub>1</sub>, …, <i>θ</i><sub><i>p</i></sub>) is continuous, twice differentiable in <b><i>θ</i></b> and monotone in components of <b><i>θ</i></b>. Following Gutenbrunner and Jurečková (1992) and Jurečková …