Evaluating the Effectiveness of Continuity Correction in Discrete Probability Distributions

Continuity correction is a very important statistical method that provides a more and more accurate estimation of a discrete probability distribution using continuous distributions. The most usual distributions in which this takes place are the binomial and Poisson distributions. This proposal seeks to provide a systematic evaluation of the effects of continuity correction on probability estimates under varying conditions. Then, five numerical examples are given to illustrate the following applications: Continuity correction to approximate probabilities with the binomial distribution, applying it to the Poisson distribution for modeling rare events, relevance, and importance in quality control applications, its consequence on epidemiological studies, and its applications in financial risk assessment. Each of these will be followed by theoretical calculations and simulations that demonstrate the differences in estimates of probability with and without continuity correction. The results will show the efficacy of the technique and thereby its implications for decision-making across diverse scenarios. In this way, the research study aspires to contribute to the literature on statistical methodologies by demonstrating the role of continuity correction in arriving at an exact estimate of probabilities.