Nonparametric estimation of regression models with unknown error distributions

在从生物信息学到经济学和商业等学科领域，研究人员根据回归模型做出重要决策，而这些模型的误差密度通常是未知的。该项目将产生一个新的采样过程，旨在选择带宽参数来估计非参数回归模型中的回归函数和误差密度。我们的方法具有实际意义，可用于研究在我们的经济和社会中可观察到的变量之间的关系。澳大利亚将从这个项目的成果中受益，在拟议的研究领域拥有自己的专家，提高澳大利亚在计量经济学和统计学方面的学术形象。

In discipline areas ranging from bioinformatics to economics and commerce, researchers make important decisions based on regression models, where the error density is often unknown. This project will result in a new sampling procedure that aims to choose bandwidth parameters for estimating the regression function and error density in nonparametric regression models. Our approach is of practical importance and can be used to investigate relationships between variables that are observable in our economy and community. The nation will benefit from the output of this project by having its own experts in the area of proposed research, raising Australia's academic profile in econometrics and statistics.