When Steve and his family moved to Pittsburgh in 1980, the first problem he talked to me about was inference in a 2 × 2 table (with ordered margins). I had always been interested in ordering problems, such as the distribution of order statistics and related matters. Superficially the parenthetical addition “with ordered margins” to the text description in the original problem Steve posed is only the slightest generalization of Steve’s original research problem (1968, 1970 with John Gilbert) of inference in a 2 × 2 table. That original problem stemmed from US Census work by Deming and Stephan (1940) “On a least squares adjustment of a sampled frequency table when the expected marginal totals are known.”
نظرات کاربران