ABSTRACT
Prediction problems arise naturally in life-testing experiments. This chapter considers a life-test experiment. In this case, twenty three ball bearings were placed on a life-test and the data on the number of million revolutions before failure of each of these ball bearings were observed. Natural for the experimenter to be interested in predicting the number of million revolutions before failure of the remaining three surviving ball bearings. In particular, the experimenter may be interested in predicting the very next failure or the very last failure. Doganaksoy and Balakrishnan presented another simple way of getting the Best Linear Unbiased Predictor (BLUP) simply from the tables of Best Linear Unbiased Estimators (BLUEs) of either the location parameter µ or the scale parameter s. These authors showed that the BLUEs of µ and s remain unchanged when the BLUP value of the order statistic Xn-s+1:n is used in turn as an observed value of that order statistic.
