Normal conditions for inference relations and injective models
β Scribed by Zhaohui Zhu; Xi'an Xiao; Yong Zhou; Wujia Zhu
- Book ID
- 104325850
- Publisher
- Elsevier Science
- Year
- 2003
- Tongue
- English
- Weight
- 418 KB
- Volume
- 309
- Category
- Article
- ISSN
- 0304-3975
No coin nor oath required. For personal study only.
β¦ Synopsis
Although fruitful representation results induced by some kinds of injective models, e.g., ΓΏltered, ranked and quasi-linear injective models, etc., have been established in the literature, it is still an open problem to characterize the family of all injective inference relations in terms of rules. The type of postulates appearing in recent literature seems to be unable to characterize this family. This brings up an interesting theoretical problem: What kind of injective inference relations may be characterized by existent types of postulates? This paper makes an initial step to answer this question. To this end, a notion of a normal condition is introduced, which subsumes all Horn and non-Horn conditions presented in the literature. We obtain some results on injective models generating inferences characterized by normal conditions, and show that these injective models must be speciΓΏc standard models. Moreover, for any set of injective models determined only by a structural property of preferential orders, if the family of inference relations induced by it can be characterized by normal conditions, then it must be a subset of ΓΏltered models in this circumstance. Thus, its associated inference relations satisfy the non-Horn rule disjunctive rationality.
π SIMILAR VOLUMES
## Abstract The calculations by the diagonal and the offβdiagonal hypervirial relations and by the sufficient conditions for these relations have been performed for the oneβdimensional box potential and the harmonic oscillator models. In the case of the box potential model, the variation method gav
## Abstract Multivariate timeβvarying volatility models have attracted a lot of attention in modern finance theory. We provide an empirical study of some multivariate ARCH and GARCH models that already exist in the literature and have attracted a lot of practical interest. Bayesian and classical te