讲座人简介：

梅长林，西安工程大学教授。1989年获西安交通大学概率论与数理统计硕士学位；1995年获比利时Limburgs Universitair Centrum大学生物统计硕士学位；2000年获西安交通大学应用数学博士学位；2001至2003年北京大学数学科学学院博士后。主要从事非参数回归与空间数据分析方面的研究，先后主持国家自然科学基金面上项目3项，发表论文90余篇，其中SCI/SSCI检索论文60余篇,论文他引1700余次。

讲座简介：

Autoregressive spatially varying coefficient models are a useful tool to simultaneously model spatial autocorrelation in the response variable and spatial heterogeneity in the regression relationship. Due to possible different spatial scales at which the explanatory variables operate, the coefficients may have different levels of spatial heterogeneity and cannot be efficiently estimated by a single bandwidth. To deal with such multiscale problem, a backfitting method is proposed to calibrate the model. A simulation study is conducted to assess the performance of the backfitting approach. The empirical results show that the backfitting estimation not only provides useful scale information for each explanatory variable, but also yields more accurate estimators of the coefficients.