Author : Jarrod Tuck
Publisher :
ISBN 13 :
Total Pages : 506 pages
Book Rating : 4.:/5 (431 download)
Book Synopsis Confidence Intervals Incorporating Prior Information by : Jarrod Tuck
Download or read book Confidence Intervals Incorporating Prior Information written by Jarrod Tuck and published by . This book was released on 2006 with total page 506 pages. Available in PDF, EPUB and Kindle. Book excerpt: This theis proposes a new approach to constructing a 1 - DC confidence interval in a large sample regression setting. This new approach utilises compromise decision theory to incorporate prior information about the nuisance regression parameter so that when it is true, the interval length is shorter than the standard 1 - DC confidence interval. This appraoch produces an interval called the Confidence Interval Incorporating Prior Information. In addition, this theory provides a performance bound on any 1- DC confidence interval. Compromise decision theory is then applied to find a 1 - DC confidence interval for the difference of two population means, when there is uncertain prior information that the population variances are equal. Here we show that, in terms of confidence interval expected length, there does not exist a 1 - DC confidence interval that provides a significant improvement over the Welch confidence interval whose length has been adjusted so that its minimum coverage is 1 - DC.