BOOTSTRAP CONFIDENCE-INTERVALS FOR ESTIMATING AUDIT VALUE FROM SKEWED POPULATIONS AND SMALL SAMPLES

Classical estimation procedures that are often used in estimating the population audit value rely on the assumption of normality. Empirical evidence suggests however, that this may not be a valid assumption. Consequently, the use of normal theory methods may lead to erroneous or misleading conclusions. The bootstrap method is an effective alternative in many cases where the classical assumptions are in question. The bootstrap method replaces complex analytical techniques by computer intensive, simulation based, empirical analysis. This study illustrates the use of the bootstrap method in estimating the audit value from skewed populations and small samples. The results of Monte-Carlo simulations indicate that the bootstrap is more effective and efficient than the normal theory method.