Simple and accurate one-sided inference from signed roots of likelihood ratios

The authors propose two methods based on the signed root of the likelihood ratio statistic for one-sided testing of a simple null hypothesis about a scalar parameter in the presence of nuisance parameters. Both methods are third-order accurate and utilise simulation to avoid the need for onerous analytical calculations characteristic of competing saddlepoint procedures. Moreover, the new methods do not require specification of ancillary statistics. The methods respect the conditioning associated with similar tests up to an error of third order, and conditioning on ancillary statistics to an error of second order. Les auteurs proposent deux methodes permettant, a partir de la racine signee du rapport des vraisemblances, d'effectuer un test unilateral d'une hypothese nulle simple sur un parametre d'echelle, en presence de parametres nuisibles. Par le biais de simulations, ces methodes permettent d'obtenir une precision du troisieme ordre tout en evitant les calculs analytiques lourds caracteristiques des methodes de point de selle concurrentes; elles n'exigent pas non plus la specification de statistiques libres. Les methodes proposees respectent, au troisieme ordre, le conditionnement associe aux tests similaires, ainsi que celui sur les statistiques libres, mais au deuxieme ordre seulement.

[1]  O. Barndorff-Nielsen On a formula for the distribution of the maximum likelihood estimator , 1983 .

[2]  Gordon Johnston,et al.  Statistical Models and Methods for Lifetime Data , 2003, Technometrics.

[3]  S. R. Chamberlin,et al.  Stable and invariant adjusted directed likelihoods , 1994 .

[4]  Thomas J. DiCiccio,et al.  Information Bias and Adjusted Profile Likelihoods , 1996 .

[5]  O. Barndorff-Nielsen,et al.  Adjusted Versions of Profile Likelihood and Directed Likelihood, and Extended Likelihood , 1994 .

[6]  Jens Ledet Jensen,et al.  Similar tests and the standardized log likelihood ratio statistic , 1986 .

[7]  Peter J. Bickel,et al.  A Decomposition for the Likelihood Ratio Statistic and the Bartlett Correction--A Bayesian Argument , 1990 .

[8]  Thomas J. DiCiccio,et al.  Simple Modifications for Signed Roots of Likelihood Ratio Statistics , 1993 .

[9]  O. E. Barndorff-Nielsen,et al.  Infereni on full or partial parameters based on the standardized signed log likelihood ratio , 1986 .

[10]  Constructing Approximately Standard Normal Pivots from Signed Roots of Adjusted Likelihood Ratio Statistics , 1994 .

[11]  Steven Stern,et al.  A Second‐order Adjustment to the Profile Likelihood in the Case of a Multidimensional Parameter of Interest , 1997 .

[12]  N. Reid,et al.  Exponential Linear Models: A Two‐Pass Procedure for Saddlepoint Approximation , 1991 .

[13]  Thomas J. DiCiccio,et al.  Approximations of marginal tail probabilities and inference for scalar parameters , 1990 .

[14]  Jerald F. Lawless,et al.  Statistical Models and Methods for Lifetime Data. , 1983 .

[15]  I. Skovgaard,et al.  An explicit large-deviation approximation to one-parameter tests , 1996 .

[16]  Nancy Reid,et al.  A simple general formula for tail probabilities for frequentist and Bayesian inference , 1999 .

[17]  B. Efron Bootstrap confidence intervals for a class of parametric problems , 1985 .

[18]  D. Lawley A GENERAL METHOD FOR APPROXIMATING TO THE DISTRIBUTION OF LIKELIHOOD RATIO CRITERIA , 1956 .

[19]  David R. Cox,et al.  Edgeworth and Saddle‐Point Approximations with Statistical Applications , 1979 .

[20]  Robert Tibshirani,et al.  A Simple Method for the Adjustment of Profile Likelihoods , 1990 .

[21]  Peter McCullagh,et al.  Conditional inference and Cauchy models , 1992 .

[22]  D. Cox,et al.  Parameter Orthogonality and Approximate Conditional Inference , 1987 .

[23]  T. Rao,et al.  Tensor Methods in Statistics , 1989 .

[24]  J. L. Jensen The modified signed likelihood statistic and saddlepoint approximations , 1992 .

[25]  Ib M. Skovgaard,et al.  Saddlepoint expansions for conditional distributions , 1987, Journal of Applied Probability.

[26]  K. Abromeit Music Received , 2023, Notes.

[27]  T. DiCiccio,et al.  Approximations of marginal tail probabilities for a class of smooth functions with applications to Bayesian and conditional inference , 1991 .

[28]  M Zelen,et al.  Estimation of exponential survival probabilities with concomitant information. , 1965, Biometrics.