Prediction Intervals for Correlated Samples.
NAVAL POSTGRADUATE SCHOOL MONTEREY CALIF
Pagination or Media Count:
In this thesis the author extends several results on prediction intervals that were obtained under the assumption that the sample observations are independent and identically distributed as Nmu,chi square. It is assumed that the samples are correlated with a prescribed correlation structure and show that many of the results available for the independent case apply equally well for the correlated samples. The correlation structure assumed occurs in variance components models in Analysis of Variance and the results can also be applied to the case where the samples have an intraclass correlation equicorrelated samples. Author
- Statistics and Probability