{"title":"指数族的 E 值:一般情况","authors":"Yunda Hao, Peter Grünwald","doi":"arxiv-2409.11134","DOIUrl":null,"url":null,"abstract":"We analyze common types of e-variables and e-processes for composite\nexponential family nulls: the optimal e-variable based on the reverse\ninformation projection (RIPr), the conditional (COND) e-variable, and the\nuniversal inference (UI) and sequen\\-tialized RIPr e-processes. We characterize\nthe RIPr prior for simple and Bayes-mixture based alternatives, either\nprecisely (for Gaussian nulls and alternatives) or in an approximate sense\n(general exponential families). We provide conditions under which the RIPr\ne-variable is (again exactly vs. approximately) equal to the COND e-variable.\nBased on these and other interrelations which we establish, we determine the\ne-power of the four e-statistics as a function of sample size, exactly for\nGaussian and up to $o(1)$ in general. For $d$-dimensional null and alternative,\nthe e-power of UI tends to be smaller by a term of $(d/2) \\log n + O(1)$ than\nthat of the COND e-variable, which is the clear winner.","PeriodicalId":501425,"journal":{"name":"arXiv - STAT - Methodology","volume":"16 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"E-Values for Exponential Families: the General Case\",\"authors\":\"Yunda Hao, Peter Grünwald\",\"doi\":\"arxiv-2409.11134\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"We analyze common types of e-variables and e-processes for composite\\nexponential family nulls: the optimal e-variable based on the reverse\\ninformation projection (RIPr), the conditional (COND) e-variable, and the\\nuniversal inference (UI) and sequen\\\\-tialized RIPr e-processes. We characterize\\nthe RIPr prior for simple and Bayes-mixture based alternatives, either\\nprecisely (for Gaussian nulls and alternatives) or in an approximate sense\\n(general exponential families). We provide conditions under which the RIPr\\ne-variable is (again exactly vs. approximately) equal to the COND e-variable.\\nBased on these and other interrelations which we establish, we determine the\\ne-power of the four e-statistics as a function of sample size, exactly for\\nGaussian and up to $o(1)$ in general. For $d$-dimensional null and alternative,\\nthe e-power of UI tends to be smaller by a term of $(d/2) \\\\log n + O(1)$ than\\nthat of the COND e-variable, which is the clear winner.\",\"PeriodicalId\":501425,\"journal\":{\"name\":\"arXiv - STAT - Methodology\",\"volume\":\"16 1\",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-09-17\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - STAT - Methodology\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2409.11134\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - STAT - Methodology","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2409.11134","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
E-Values for Exponential Families: the General Case
We analyze common types of e-variables and e-processes for composite
exponential family nulls: the optimal e-variable based on the reverse
information projection (RIPr), the conditional (COND) e-variable, and the
universal inference (UI) and sequen\-tialized RIPr e-processes. We characterize
the RIPr prior for simple and Bayes-mixture based alternatives, either
precisely (for Gaussian nulls and alternatives) or in an approximate sense
(general exponential families). We provide conditions under which the RIPr
e-variable is (again exactly vs. approximately) equal to the COND e-variable.
Based on these and other interrelations which we establish, we determine the
e-power of the four e-statistics as a function of sample size, exactly for
Gaussian and up to $o(1)$ in general. For $d$-dimensional null and alternative,
the e-power of UI tends to be smaller by a term of $(d/2) \log n + O(1)$ than
that of the COND e-variable, which is the clear winner.