这篇文章的那段原文是这样的，我摘录出来
The story proceeds as follows. We first introduce the inputs to time-series regressions: the explanatory variables and the returns to be explained(sections 2 and 3). We then use the regressions to attack our two central asset-pricing issues: how do different combinations of variables captures (a) the common variation through time in the returns on bonds and stocks(section 4) and (b) the cross-section of average returns(section 5).
主要说的是用B/M和SIZE这两种变量构成指标体系形成完整的收益率中的风险因素
B/M和SIZE分别用五个级别来划分，这样在直角坐标系下面就形成了25个区域，这25个区域的风险都不尽相同

应该是先按size排序五等分再将每一份按b/m再重新排序再次五等分就是25组