Empirical process results for exchangeable arrays

Exchangeable arrays are natural tools to model common forms of dependence between units a sample. Jointly exchangeable well suited dyadic data, where observed random variables indexed by two from the same population. Examples include trade flows countries or relationships in network. Separately multiway clustering, sharing cluster (e.g., geographical areas sectors activity when considering individual wages) may be dependent an unrestricted way. We prove uniform laws large numbers and central limit theorems for such arrays. obtain these results under moment restrictions conditions on class functions as those typically assumed with i.i.d. data. also show convergence bootstrap processes adapted