Two leading estimators of the finite population distribution function are studied. Their large-sample mean squared errors are compared, under a given model. Neither estimator is universally better; circumstances are identified where the model-based estimator would do poorly, despite employing the correct model.
No takes yet. Share an insight, caveat, or question.
Chambers et al. (1992) studied this question.
Synapse has enriched one closely related paper. Consider it for comparative context: