Identification-robust nonparametric inference in a linear IV model

For a linear IV regression, we propose two new inference procedures on parameters of endogenous variables that are robust to any identification pattern, do not rely first-stage equation, and account for heteroskedasticity unknown form. Building Bierens (1982), first an Integrated Conditional Moment (ICM) type statistic constructed by setting the value under null hypothesis. The ICM procedure tests at same time coefficient specification model. We then adopt conditionality principle condition set statistics informs strength. Our uniformly control size irrespective They powerful nonlinear form link between instruments competitive with existing in simulations application.