A1101
Title: Acceptability criteria for randomizations: A generalized definition of randomized designs
Authors: Tirthankar Dasgupta - Rutgers University (United States) [presenting]
Abstract: The focus is on how different randomized designs in classical statistics can be viewed as special cases of a randomized design based on a class of general criteria for acceptance of treatment assignments to a finite population of experimental units. It is argued that this view results in a unified perspective of randomized experiments and leads to an intuitive, model-free inference for causal effects of treatments.
