中文摘要：

变系数模型是线性模型的有用推广，它允许回归系数是某个变量的函数，近年来在统计分析中得到广泛的应用．文中研究回归变量都是随机时的变系数模型，提出运用小波的方法估计变系数模型中的函数系数，并在较弱的条件下得到了变系数模型小波估计的渐近正态性．

英文摘要：

Varying-coefficient model is a useful extension of the classical linear model. Due to its significant characteristics that the regression coefficients can change over certain covariates, recently it has been widely applied in all kinds of statistic analysis. In this article, a wavelet procedure is proposed to construct the estimators of functional coefficients of varying-coefficient model in which all covariates are random. The asymptotic normalities of the proposed estimators are also established under weaker conditions.