A1301
Title: Aspects of statistical testing with incomplete observations
Authors: Daniel Gaigall - FH Aachen University of Applied Sciences (Germany) [presenting]
Philipp Wuebbolding - FH Aachen University of Applied Sciences (Germany)
Abstract: Statistical testing based on incomplete observations is considered for various testing problems including goodness-of-fit, symmetry, homogeneity, and independence. A specific structure of incomplete data is identified and utilized to enable the analysis of statistical procedures using suitable results such as laws of large numbers and central limit theorems. A general and novel concept for testing in incomplete data situations is developed. An example application involves testing goodness-of-fit for normality. Handling these situations requires the development of new methods for determining critical values. Simulation experiments demonstrate the potential of the approach.
