Key points are not available for this paper at this time.
Monte Carlo simulation studies are numerical methods for conducting computer experiments based on generating pseudo-random observations from a known truth. Monte Carlo simulation studies -referred from now on as simulation studies for conciseness -represent a powerful tool and have several practical applications in statistical and biostatistical research: among others, evaluating new or existing statistical methods, comparing them, assessing the impact of modelling assumption violations, and helping with the understanding of statistical concepts. Establishing properties of current methods is necessary to allow using them with confidence; however, sometimes properties are very hard (if not impossible) to derive analytically: large sample approximation is possible, but evaluating the goodness of the approximation to finite samples is required. Approximations often require assumptions as well: what are the consequences of violating such assumptions? Simulation studies can help answer these questions. They can also help answer additional questions such as: is an estimator biased in a finite sample? Do confidence intervals for a given parameter achieve the desired nominal level of coverage? How does a newly developed method compare to an established one? What is the power to detect a desired effect size under complex experimental settings and analysis methods?
Alessandro Gasparini (Wed,) studied this question.