ASYMPTOTIC CUMULANTS OF SOME INFORMATION CRITERIA
نویسندگان
چکیده
منابع مشابه
Asymptotic expansions of moments and cumulants
Many parameters may be expanded as series with terms involving products of expectations and their estimates expanded as series involving products of averages. The computation of moments and cu-mulants of such estimates may be organized if the terms of the series expansions of parameters, estimates and their moments are considered as functions applied to lists. The lists form a vectors space ass...
متن کاملHigher Order Asymptotic Cumulants of Studentized Estimators in Covariance Structures
This note gives corrections of the numerical values of the higher-order asymptotic variances (standard errors) and kurtoses of the Studentized estimators (and consequently the root mean square errors of the two-term Edgeworth expansions) shown in the tables of Ogasawara (2008a, 2008b). The corrections are due to a minor bug in the computer program. The corrected values given in the following Ta...
متن کاملAsymptotic Normality Through Factorial Cumulants and Partition Identities
In the paper we develop an approach to asymptotic normality through factorial cumulants. Factorial cumulants arise in the same manner from factorial moments as do (ordinary) cumulants from (ordinary) moments. Another tool we exploit is a new identity for 'moments' of partitions of numbers. The general limiting result is then used to (re-)derive asymptotic normality for several models including ...
متن کاملSome Information Theoretic Optimality Criteria for General Classification'
Central to the various purposes for which general classification is practiced is the preservation, or summarizing, of information (Michener, 1963; Sneath, 1957; Sokal, 1962; Farris, 1967; Johnson, 1968; Rogers, 1961; Wirth, Estabrook, and Rogers, 1966). This discussion will present one method in which this fundamental concept can be made more precise and, within that context, suggest possible l...
متن کاملSome Criteria of Univalence1
is the Schwarzian derivative of w =/(z). The two cases treated in [3 ] were m(\z\) =ir2/2 and m(\z\) =2(1 — \z\ 2)~2. The constants appearing in both criteria are the largest possible. In the first case this is shown by the existence of the nonunivalent function w = tan ir( 1 + e)z/2 (e>0) for which {w, z\ =7r2(l+e)2/2, and in the second case by an example constructed by E. Hille [2]. Other cri...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Journal of the Japanese Society of Computational Statistics
سال: 2016
ISSN: 0915-2350,1881-1337
DOI: 10.5183/jjscs.1512001_225