论文部分内容阅读
In this paper，the empirical likelihood confidence regions for the regression coefficient in a linear model are constructed under m-dependent errors. It is shown that the blockwise empirical likelihood is a good way to deal with dependent samples.