Executive Summary : | Inference upon reliability characteristics in presence of truncated observations requires development of different efficient statistical procedures to derive results which are useful in practical studies. Most of the inferences in this context are derived when exact lifetimes of various test units may be partially known to the experimenter. One of the important reasons for this is that there is frequent need to analyze empirical data before lifetimes of all units are recorded. Often in such studies inference upon unknown parametric functions and predictive inference of future observable are required. Deriving estimates improving benchmark estimators are also of significant important in lifetime analysis. Such investigations commonly require useful statistical/mathematical analysis. In this research proposal we plan to derive various parametric and predictive inferences for problems having applications in life test studies. Further inference upon stress-strength parameters plays important role in real life applications. Also many systems or subsystems fail due to more than one mode of failure. In many applications it is important to distinguish between modes of failure to improve reliability of components. Thus competing risk models have found wide applications in life test studies. We mainly focus to derive theoretical procedures to estimate parametric functions of families of probability distributions. We will further examine the behavior of proposed procedures empirically as well. In this research proposal we plan to work on inference problems related to these practical fields of study. |