Confidence intervals on linear combinations of variance components that are unrestricted in sign
- 1 April 1990
- journal article
- research article
- Published by Taylor & Francis in Journal of Statistical Computation and Simulation
- Vol. 35 (3-4) , 135-143
- https://doi.org/10.1080/00949659008811240
Abstract
A new method is proposed for constructing confidence intervals on where the sign of γ is unrestricted and unknown, and are independently distributed chi-squared random variables with ni degrees of freedom for Computer simulation is used to illustrate that the method provides a confidence coefficient that is generally close to the stated level. The method is illustrated by using it to test a main effect in a random three-factor crossed design.Keywords
This publication has 5 references indexed in Scilit:
- One-sided confidence intervals on nonnegative sums of variance componentsStatistics & Probability Letters, 1989
- Confidence intervals on a difference of expected mean squaresJournal of Statistical Planning and Inference, 1988
- Confidence Intervals on Nonnegative Linear Combinations of VariancesJournal of the American Statistical Association, 1980
- Confidence Intervals on Variance Components in Three-Factor Cross-Classification ModelsTechnometrics, 1980
- Approximate Confidence Limits on the Mean of X + Y Where X and Y are Two Tabled Independent Random VariablesJournal of the American Statistical Association, 1974