基于复合分位数的半参数模型平均边际回归

his paper considers the composite quantile regression for semiparametric model averaging marginal regression. Firstly,this paperuses semiparametric model averaging of marginal regression based on local linear composite quantile regression to estimate the marginal regression function. Moreover, this paper estimates and selects weights for one-dimensional marginal regression models by combining penalized composite quantile regression with an adaptive LASSO penalty function, which is then used to estimate the joint regression function. This paper further demonstrates the asymptotic normality of the estimator and the sparsity of the weight estimates under certain conditions. Finally, this paper shows the superior performance of our proposed method through numerical simulations.