``Best possible'' upper bounds for the first positive zeros of Bessel functions -- the finite part (Q2564268): Difference between revisions

The authors show that the first three terms in an expansion of \textit{F. W. J. Olver} [Proc. Cambridge Philos. Soc. 47, 699-712 (1951; Zbl 0054.03301)] for the zeros \(j_{\nu k}\) of the Bessel function \(J_\nu\) provide an upper bound for \(j_{\nu k}\), \(k\in\{1,2,3\}\), \(0<\nu\leq 10\), and that the first two terms in the same expansion provide a lower bound for \(j_{\nu k}\), \(k\in\{2,3\}\), \(0\leq\nu\leq 10\). This extends previous results by \textit{H. W. Hethcote} [SIAM J. Math. Anal. 1, 147-152 (1970; Zbl 0199.49902)] for \(k=1\), \(0\leq \nu<\infty\), and by \textit{T. Lang} and \textit{R. Wong} [J. Comput. Appl. Math. 71, No. 2, 311-329 (1996)] for the upper bound with \(k\in\{1,2\}\), \(\nu>10\).