Abstract
A general expression for the UMP unbiased testsĀ of the hypotheses \(H_1:\theta \le \theta _0\) and \(H_4:\theta =\theta _0\) in the exponential family.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
Notes
- 1.
This problem is discussed in Section 3 of Hodges andĀ Lehmann (1954).
- 2.
The relationship \(W=Y/(1+Y)\) shows the F and beta distributions to be equivalent. Tables of these distributions are discussed in Chapters 24 and 26 ofĀ Johnson, Kotz and Balakrishnan (1995. Vol. 2). Critical values of F are tabledĀ by Mardia and Zemroch (1978), who also provide algorithms for the associated computations.
- 3.
A comparison of these limits with those obtained from the equal-tails test is given byĀ ScheffĆ© (1942); some values of \(C_1\) and \(C_2\) are provided by Ramachandran (1958).
- 4.
The literature on regression is enormous and we treat the simplest model. Some texts on the subjectĀ include Weisberg (1985),Ā Atkinson and Riani (2000) and Ā Chatterjee Hadi and Price (2000).
- 5.
A method for obtaining the size of this effect was developed by Neyman, and tables have been computed on its basis by Fix. This work is reportedĀ by Bennett (1957).
- 6.
- 7.
They also do not occur when the posterior distribution of \(\Theta \) is discrete.
- 8.
For a closely relatedĀ result. see OdĆ©n and Wedel (1975).
- 9.
The problem is simplified here, not just due to the assumption of normality, but also by the assumption that the distribution of \(X_i - u_i\) does not depend on \(\mu _i\), and similarly for the \(Y_i\). In general, when observations are paired approximately according to covariates, then pairs cannot be treated as if they were sampled from a population. A recent treatment is provided in Bai, Romano, and Shaikh (2019).
- 10.
See, for example, Billingsley (1995),Ā p. 417.
- 11.
Actually, all that is needed is that \(f_1,\ldots ,f_c\in \mathcal{F}\), where \(\mathcal F\) is any family containing all normal distributions.
- 12.
A systematic account of this distribution can be foundĀ in in Owen (1985)Ā and Johnson Kotz and Balakrishnan (1995).
- 13.
For additional information concerning inference in inverse Gaussian distributions, see Ā Folks and Chhikara (1978) and Ā Johnson, Kotz and Balakrishnan (1994, volume 1).
- 14.
A similar conclusion holds in the problem of constructing a confidence interval for the ratio of normal meansĀ (Fiellerās problem), as discussed inĀ Koschat (1987). For problems where it is impossible to construct confidence intervals with finite expected length , see Gleser and Hwang (1987).
- 15.
For the corresponding result concerning one-sided confidence bounds, see Madansky (1962).
- 16.
The distribution of R is reviewed byĀ Johnson and Kotz (1970, Vol. 2, Section 32) and Ā Patel and Read (1982).
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
Copyright information
Ā© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this chapter
Cite this chapter
Lehmann, E.L., Romano, J.P. (2022). Unbiasedness: Applications to Normal Distributions; Confidence Intervals. In: Testing Statistical Hypotheses. Springer Texts in Statistics. Springer, Cham. https://doi.org/10.1007/978-3-030-70578-7_5
Download citation
DOI: https://doi.org/10.1007/978-3-030-70578-7_5
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-70577-0
Online ISBN: 978-3-030-70578-7
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)