Continuity corrections for integer-valued saddlepoint approximations

DOI10.1016/J.SPL.2006.03.005MaRDI QIDQ2497807zbMATH OpenOpenAlexFDO

Authors Thomas Fung, John Robinson

Publication date 4 August 2006

Published in Statistics \& Probability Letters (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1016/j.spl.2006.03.005

zbMATH Keywords

continuity correctionsLugannani-Rice formularelative errorssaddlepoint approximationtail area approximations

Mathematics Subject Classification ID

Asymptotic approximations, asymptotic expansions (steepest descent, etc.) (41A60) Large deviations (60F10) Asymptotic distribution theory in statistics (62E20)

Review text

Let \(X_1,X_2,\dots\), be integer-valued r.v's taking values \(N_1, N_1+1,\dots, N_2\), where \(N_1\), \(N_2\) can be infinite. The authors discuss the accuracy of the saddlepoint approximations for the upper tail \(P(\overline{X}\geq x)\).

Recommendations

• Saddlepoint approximation for the distribution function near the mean
• Saddlepoint approximation to the distribution of general statistics
• A note on the tail accuracy of the univariate saddlepoint approximation
• Tail Probability Approximations
• Approximate and estimated saddlepoint approximations

Cites work

• Saddlepoint Approximations in Statistics
• Tail Probability Approximations

Cited in

(1)

• 2008 Pitman Medal

This page was built for publication: Continuity corrections for integer-valued saddlepoint approximations

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2497807)