Choosing between Normal and Poisson Approximations to the Cumulative Binomial Probabilities: A New Method | ||||
| The Egyptian Statistical Journal | ||||
| Article 11, Volume 35, Issue 1, June 1991, Page 125-143 PDF (7.92 MB) | ||||
| Document Type: Original Article | ||||
| DOI: 10.21608/esju.1991.315001 | ||||
 View on SCiNiTO | ||||
| Authors | ||||
| Abdul-Mordy H. Azzam* ; Adel M. Halawa; Osama A. Hussien | ||||
| Dep.of Statistics,, Faculty of Commerce, Alexandria University, Alexandria, Egypt | ||||
| Abstract | ||||
| The problem of approximating the exact cumulative binomial probabilities has received a great deal of attention in the literature. Several methods of approximation were suggested. Most of these methods fall in one of two broad classes: the class of normal approximations or the class of Poison approximations. The main goal of this paper is giving a new method which helps the researcher to determine the class he may use. Extensive computer runs show that this method dominates what are known as rules of thumb for the domain of applicability of the normal approximation. | ||||
| Keywords | ||||
| Class of Normal Approximations; Class of Poison Approximations; Maximum Absolute Error (MABS) | ||||
|
Statistics Article View: 82 PDF Download: 22 |
||||
