Abstract
AMONG types of statistical inference about unknown parameters statements are possible which have a statistical truth, that is, they are random variables such that within the statistical framework adopted the probability of their being in error is known. In these statements intervals, called by Neyman confidence intervals, are assigned to the value of an unknown parameter. On generalization to more than one unknown parameter these intervals become multidimensional regions; but I have pointed out1 that the existence of such regions does not of itself imply in Neyman's theory the corresponding existence of regions of lower order, equivalent to the elimination of irrelevant unknown parameters.
Similar content being viewed by others
Article PDF
References
Ann. Math. Stat., 10, 129 (1939).
Sankhya, 7, 129 (1945).
Proc. Camb. Phil. Soc., 32, 560 (1936).
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
BARTLETT, M. A General Class of Confidence Interval. Nature 158, 521 (1946). https://doi.org/10.1038/158521a0
Issue Date:
DOI: https://doi.org/10.1038/158521a0
This article is cited by
Comments
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.