Correlated or matched data is frequently collected under many study designs in applied sciences such as the social, behavioral, economic, biological, medical, epidemiologic, health, public health, and drug developmental sciences in order to have more efficient design and to control for potential confounding factors in the study. Challenges with respect to availability and cost commonly occur with matching observational or experimental study subjects, thus researchers frequently encounter situations where the observed sample consists of a combination of correlated and uncorrelated data due to missing responses (partially correlated data). Ignoring cases with missing responses, when analyzing the data, will introduce bias in the inference and lower the power of the testing procedure. This paper discusses and proposes some nonparametric testing procedures to handle data when partially observed correlated data is available without ignoring the cases with missing responses. Therefore, we will introduce more powerful testing procedures which combine all cases in the study. A theoretical and numerical investigation will be provided. The proposed testing procedures will be applied to simulated and genetic data. This is an important research area and solving one of the challenging issues in missing data problem will enrich the research in this area and help the researcher to have more powerful test procedure.
Available at: http://works.bepress.com/robert_vogel/72/