Summary
The estimation of the regression coefficient of a population, defined byE (y)= γ+βx, incorporating two preliminary tests of significance has been discussed. The experimenter has two random samples of different sizes from two such populations, as defined above, with regression coefficientsβ 1 andβ 2 respectively, whereβ 2 may possibly be equal toβ 1. Besides this, it is also conjectured that the common conditional varianceσ 2 of the two populations has a specified valueσ 20 . The two preliminary tests are used to resolve these two uncertainties.
Similar content being viewed by others
References
Bancroft, T.A.: On biases in estimation due to the use of preliminary tests of significance. Ann. Maths. Stats.15, 1944, 190–204.
—: Analysis and inference for incompletely specified models involving the use of preliminary test(s) of significance. Biometrics20, 1964, 427–442.
Han, C.P., andT.A. Bancroft: On pooling means when variance is unknown. Jour. Ame. Stats. Assoc.63, 1968, 1333–1342.
Kale, B.K., andT.A. Bancroft: Inferences for some incompletely specified models involving normal approximations to discrete data. Biometrics23, 1967, 335–348.
Mosteller, F.: On pooling data. Jour. Ame. Stats. Assoc.43, 1948, 231–242.
Shewhart, W.A.: Economic control of quality of manufactures product. New York 1931.
Shukla, N.D.: Unpublished Ph.D. Thesis. Lucknow, India, 1974.
Shukla, N.D., andS.R. Srivastava: Pooling of regression coefficients when conditional variances are unknown. Trabajos De Estadistika24, 1973, 141–147.
Author information
Authors and Affiliations
Additional information
The author has rejoined Lucknow University, Lucknow India on Oct. 4, 1976 after availing leave for two years.
Rights and permissions
About this article
Cite this article
Shukla, N.D. Estimation of a regression coefficient after two preliminary tests of significance. Metrika 26, 183–193 (1979). https://doi.org/10.1007/BF01893486
Received:
Revised:
Issue Date:
DOI: https://doi.org/10.1007/BF01893486