Abstract
A systematic method of the construction of a confidence interval for the difference between two means is given in the exponential and gamma cases. The application of a similar method to the Behrens-Fisher type problem is also given. Further, the numerical calculation of coverage probabilities is done and a comparison of the confidence interval proposed in this paper with that based on the Fisher-Welch-Wald similar test is given. As a result, the method is seen to be reasonable.
Similar content being viewed by others
References
Thursby, J. G. (1992). A comparison of several exact and approximate tests for structural shift under heteroscedasticity. J. Econometrics 53, 363–386.
Tsui, K. and Weerahandi, S. (1989). Generalized p values in significance testing of hypotheses in the presence nuisance parameters. J. Amer. Statist. Assoc., 84, 602–607.
Weerahandi, S. (1987). Testing regression equality with unequal variances. Econometrica 55, 1211–1215.
Weerahandi, S. (1993). Generalized confidence intervals. J. Amer. Statist. Assoc., 88, 899–905.
Weerahandi, S. (1995). Exact Statistical Methods for Data Analysis. Springer, New York.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Akahira, M. Confidence intervals for the difference of means: application to the Behrens-Fisher type problem. Statistical Papers 43, 273–284 (2002). https://doi.org/10.1007/s00362-002-0100-4
Received:
Revised:
Issue Date:
DOI: https://doi.org/10.1007/s00362-002-0100-4