在應用統計中，使用二項分配近似常態分配使用連續性校正是非常有用的。首先，論文的第一部分，回顧二項分配和中央極限定理。如果樣本數較大，二項分配會近似於常態分配。連續性校正是為了進一步提高二項分配近似常態分配的準確，其中較廣為人知的連續性較正為Yates’s correction for continuity (Yates, 1934; Cox, 1970)。此外我們還介紹比較少人知道的Cressie’s finely tuned continuity correction (Cressie, 1978)。我們將連續性較正應用在統計製程的問題中。此外，我們進行數值模擬研究，比較Yates’s correction for continuity跟Cressie’s finely tuned continuity correction。;In applied statistics, the continuity correction is useful when the binomial distribution is approximated by the normal distribution. In the first part of this thesis, we review the binomial distribution and the central limit theorem. If the sample size gets larger, the binomial distribution approaches to the normal distribution. The continuity correction is an adjustment that is made to further improve the normal approximation, also known as Yates’s correction for continuity (Yates, 1934; Cox, 1970). We also introduce Cressie’s finely tuned continuity correction (Cressie, 1978), which are less known for statisticians. We discuss the application of these continuity corrections to the problem of statistical process control and confidence limit. In addition, we perform numerical studies to compare these corrections.
