Instrumental variable estimators can estimate a local average treatment effect i.e. the impact of the treatment only for the group affected by the instrument provided one has a binary treatment (or endogenous regressor) and binary instrument. In other settings, one still recovers a consistent parameter estimate of the endogeneous regressor but it can only take a causal meaning if treatment effects are constant.

看到哭都看不懂.

不好意思, endogenous regressor我想应该指的是DEPENDENT VARIABLE,对吗?能帮我看看这段话吗? 谢谢