Recently Gupta and Gupta  have discussed point estimation for R = Pr(a' ‘ b’y) where xpx| and yqx1are two non-independent multivariate normal variates and a and b are two known vectors. We show how a lower confidence bound can be obtained for R and how this can be used to refine the inference for the application considered by Gupta and Gupta. Furthermore the case of independent x and y is considered and an application of this case is presented.
- confidence bounds
- stress-strength models
ASJC Scopus subject areas
- Statistics and Probability
- Statistics, Probability and Uncertainty