A Databased Measure for Statistical Inference

• Activity Number: 297
• Type: Contributed
• Date/Time: Tuesday, July 31, 2012 : 8:30 AM to 10:20 AM
• Sponsor: Biometrics Section
• Author(s): Walid Nuri*+
• Keywords: simulation ; supporting event ; population image ; measure of favorability ; optimal estimator

Abstract

This paper presents an alternative to the probabilistic approach in measuring events of a sample space. Most of our statistical inference methods are based on mathematical idealizations using models to describe the information found through random samples from a particular population. This method shows that statistical inference rules can be created directly from this information without going further to study a population through probabilistic models. It introduces the concept of a population image created using the measurements of a given random sample (selected for inference about a population parameter). This approach employs a computer simulation technique to select (with replacement) a very large number m, say, of random samples from the measurements. Then, for an event E in a sample space, the function T is defined as the proportion of the number of cases within these samples that