View : 825 Download: 0

Full metadata record

DC Field Value Language
dc.contributor.author장연정-
dc.creator장연정-
dc.date.accessioned2016-08-26T02:08:25Z-
dc.date.available2016-08-26T02:08:25Z-
dc.date.issued1999-
dc.identifier.otherOAK-000000002188-
dc.identifier.urihttps://dspace.ewha.ac.kr/handle/2015.oak/192468-
dc.identifier.urihttp://dcollection.ewha.ac.kr/jsp/common/DcLoOrgPer.jsp?sItemId=000000002188-
dc.description.abstract지금까지 이항모수로 대표되는 비율에 대한 신뢰구간으로 Wald 신뢰구간, Clopper-Pearson exact 신뢰구간, score 신뢰구간, adjusted Wald 신뢰구간 등등 여러 가지 방법들이 제안되었다. 많은 기초통계학 책을 통해 잘 알려져 있는 Wald 신뢰구간은 coverage probability의 측면에서 단점을 가지고 있다. 정확한 coverage probability를 구하고자 한다면 flopper-Pearson exact 신뢰구간을 이용할 수 있겠지만 공식이 계산하기 쉽지 않다. Score 신뢰구간은 coverage probability로 보면 Wald 신뢰구간보다 더 낫지만 역시 공식이 다소 복잡하다. Coverage probability와 계산의 간편성이란 두 가지 측면을 함께 고려한다면 adjusted Wald 신뢰구간이 바람직하다고 할 수 있다. 이 논문에서 우리는 coverage probability를 근거로 Monte-Carlo simulation을 통해 위에서 언급한 네 가지 신뢰구간에 대한 비교를 하였고, adjusted Wald 신뢰구간이 상당한 효용성을 갖고 있음을 보인다. ; There exist several ways suggested so far to construct a confidence interval for a proportion which is represented by a binomial parameter: Wald confidence interval, and so on. The Wald confidence interval is well-known and appears in many introductory text books. But it has a deficiency in view of the coverage probability. The Clopper-Pearson exact confidence interval seems to be better if one wishes an exact coverage probability, but its formula is not so compute. The score confidence interval is rather better than the Wald confidence interval in view of coverage probabilities, but its formula is still difficult to compute. In regards of both the coverage probabilities and the ease of computation, the adjusted Wald confidence interval is recommendable. In this Thesis we compare the four methods mentioned above based on their coverage probabilities by Monte-Carlo simulation.-
dc.description.tableofcontentsABSTRACT CHAPTERS 1. INTRODUCTION = 1 2. CONfIDENCE INTERVALS FOR BINOMIAL PARAMETER p = 4 3. COVERAGE PROBABILITIES = 11 4. DISCUSSIONS = 17 REFERENCES = 19 APPENDICES A. PROGRAMS FOR SIMULATION = 21 B. TABLES FOR CHAPTER 3 = 28 C. FIGURES FOR CHAPTER 3 = 33 논문초록 = 42 감사의 글-
dc.formatapplication/pdf-
dc.format.extent1033649 bytes-
dc.languageeng-
dc.publisher이화여자대학교 대학원-
dc.titleOn the coverage probabilities of confidence interval estimation for binomial proportions-
dc.typeMaster's Thesis-
dc.identifier.thesisdegreeMaster-
dc.identifier.major대학원 통계학과-
dc.date.awarded1999. 8-
Appears in Collections:
일반대학원 > 통계학과 > Theses_Master
Files in This Item:
There are no files associated with this item.
Export
RIS (EndNote)
XLS (Excel)
XML


qrcode

BROWSE