scientific article; zbMATH DE number 3967665
[For the entire collection see Zbl 0564.00024.] Results of \textit{R. A. Johnson} and \textit{G. G. Roussas} [Ann. Math. Stat. 40, 1207-1215 (1969; Zbl 0188.514); ibid. 41, 918-938 (1970; Zbl 0205.199)] and the author [Bull. Soc. Math. Grèce, n. Ser. 14, 98-121 (1973; Zbl 0295.62020); ibid. 15, 170-180 (1974; Zbl 0324.62018)] are specialized to the i.i.d. case, and they are found to be applicable to the logarithmic, logistic and Cauchy distributions giving asymptotically optimal tests for testing \(H_ 0: \theta =\theta_ 0\) against one-sided and two-sided alternatives. The verification of the regularity conditions is facilitated by a proposition on derivatives in quadratic mean and a proposition on convergence in probability to \(\pm \infty\). It is noted that Wald's theory does not provide tests in any of the distributions mentioned above, since the MLE of \(\theta\) is not available in closed form in any of them.
This page was built for publication:
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3734862)