TY - JOUR
T1 - Testing the difference of two binomial proportions
T2 - Comparison of continuity corrections for saddlepoint approximation
AU - Xu, Lu
AU - Kolassa, John E.
N1 - Funding Information:
We gratefully acknowledge funding support from NSF DMS-0505499.
PY - 2008/1
Y1 - 2008/1
N2 - We carried out a simulation study based on the methodology of Newcombe (1998) to compare tests for the difference of two binomial proportions by applying different continuity corrections on saddlepoint approximation to tail probabilities. In this article, we proposed a new continuity correction based on the least common multiple of two sample sizes. We evaluated that the best test should have the actual Type I error rates that are, on the whole, closest to α, but not exceeding α, where α is nominal level of significance.
AB - We carried out a simulation study based on the methodology of Newcombe (1998) to compare tests for the difference of two binomial proportions by applying different continuity corrections on saddlepoint approximation to tail probabilities. In this article, we proposed a new continuity correction based on the least common multiple of two sample sizes. We evaluated that the best test should have the actual Type I error rates that are, on the whole, closest to α, but not exceeding α, where α is nominal level of significance.
KW - Continuity corrections
KW - Lugannani and Rice formula
KW - Saddlepoint approximation
KW - Type I error rate
UR - https://www.scopus.com/pages/publications/46249131128
UR - https://www.scopus.com/pages/publications/46249131128#tab=citedBy
U2 - 10.1080/03610920701875275
DO - 10.1080/03610920701875275
M3 - Article
AN - SCOPUS:46249131128
SN - 0361-0926
VL - 37
SP - 2213
EP - 2218
JO - Communications in Statistics - Theory and Methods
JF - Communications in Statistics - Theory and Methods
IS - 14
ER -