最近收到审稿意见，其中一条是针对回归系数标准误提出的：“It seems that clustering would be important herebut difficult.... I would suggest that the authors explorealternative (non-asymptotic) approaches and see if those could be used here.”我在初稿中是在个人层面针对标准误做的聚类调整。审稿人建议使用non-asymptotic approaches。

还请问大牛们，哪些标准误的调整方式属于non-asymptotic approaches？stata中如何实现？

多谢！