Robust inference policies: preliminary report

Reasoning under uncertainty often involves a great deal of judgmental imprecision. The subjective (or database-retrieved) uncertainty estimates that serve as the ingredients of an uncertainty calculus are often perceived as arbitrary, imprecise, or uncertain. One consequence of this judgmental imprecision is that many decision makers (and researchers) avoid using an explicit uncertainty calculus for fear of being subject to a garbage-in garbage-out problem. A series of Monte Carlo studies were performed to assess the extent to which different inference procedures robustly output reasonable belief values in the context of increasing levels of judgmental imprecision. It was found that, when compared with an equal-weights linear model, the Bayesian procedures are more likely to deduce strong support for a hypothesis. But the Bayesian procedures are also more likely to strongly support the wrong hypothesis. Bayesian techniques are mote powerful, but also more error prone