TY - BOOK ID - 16388433 TI - Encompassing : formulation, properties and testing PY - 1997 VL - 446 SN - 3540618902 3642468322 9783540618904 PB - Berlin : Springer, DB - UniCat KW - Statistical hypothesis testing KW - Social sciences KW - Statistical methods KW - Mathematical models KW - Statistical hypothesis testing. KW - Mathematical models. KW - Statistical methods. KW - Stochastic processes KW - 330.115 KW - 330.115 Econometrie KW - Econometrie KW - Hypothesis testing (Statistics) KW - Significance testing (Statistics) KW - Statistical significance testing KW - Testing statistical hypotheses KW - Distribution (Probability theory) KW - Hypothesis KW - Mathematical statistics KW - Economic theory. KW - Statistics . KW - Economic Theory/Quantitative Economics/Mathematical Methods. KW - Statistics for Business, Management, Economics, Finance, Insurance. KW - Statistical analysis KW - Statistical data KW - Statistical science KW - Mathematics KW - Econometrics KW - Economic theory KW - Political economy KW - Economic man KW - Social sciences - Statistical methods KW - Social sciences - Mathematical models UR - https://www.unicat.be/uniCat?func=search&query=sysid:16388433 AB - The history of many sciences is characterized by an almost continuous emer­ gence of new theories. From a normative point of view, the survival of a new theory should mainly be determined by its ability to explain a new body of facts which the existing theories are unable to explain. If in addition the new theory is able to explain all the results obtained by the existing theories and if it can point out why these theories fail to explain certain facts, it should become the dominant theory. Otherwise, it might coexist with other theories for some time. Hence, a new theory ought to be judged not only by confronting it with existing facts, but also by confronting it with existing theories. The idea that a theory should be able to account for the results ob­ tained by other theories, although implicitly adhered to by many scientists, has rarely been formalized. The statistics literature on parametric hypoth­ esis testing, though, might be seen as an instance of such a formalization. ER -