Bernoulli

  • Bernoulli
  • Volume 7, Number 3 (2001), 473-485.

Likelihood computations without Bartlett identities

Per Aslak Mykland

Full-text: Open access

Abstract

The signed square root statistic R is given by sgn( \hatθ- θ) ( l( \hatθ) - l( θ) )1/2, where l is the log-likelihood and \hatθ is the maximum likelihood estimator. The pth cumulant of R is typically of the form n-{p/2}kp + O(n-{p+2)/2) , where n is the number of observations. This paper shows how to symbolically compute kp without invoking the Bartlett identities. As an application, we show how the family of alternatives influences the coverage accuracy of R.

Article information

Source
Bernoulli, Volume 7, Number 3 (2001), 473-485.

Dates
First available in Project Euclid: 22 March 2004

Permanent link to this document
https://projecteuclid.org/euclid.bj/1080004761

Mathematical Reviews number (MathSciNet)
MR2002f:62018

Zentralblatt MATH identifier
0987.62017

Keywords
Bartlett correction convergence of cumulants unconditional accuracy

Citation

Aslak Mykland, Per. Likelihood computations without Bartlett identities. Bernoulli 7 (2001), no. 3, 473--485. https://projecteuclid.org/euclid.bj/1080004761


Export citation

References

  • [1] Andrews, D.F. and Stafford, J.E. (1993) Tools for the symbolic computation of asymptotic expansions. J. Roy. Statist. Soc. Ser. B, 55, 613-627. Abstract can also be found in the ISI/STMA publication
  • [2] Andrews, D.F. and Stafford, J.E. (2000) Symbolic Computation for Statistical Inference. Oxford: Oxford University Press.
  • [3] Barndorff-Nielsen, O.E. (1986) Inference on full or partial parameters based on the standardized signed log likelihood ratio. Biometrika, 73, 307-322.
  • [4] Barndorff-Nielsen, O.E. (1991) Modified signed log likelihood ratio. Biometrika, 78, 557-563.
  • [5] Barndorff-Nielsen, O.E. and Cox, D.R. (1989) Asymptotic Techniques for Use in Statistics. London: Chapman & Hall.
  • [6] Barndorff-Nielsen, O.E. and Wood, A.T.A. (1998) On large deviations and choice of ancillary for p* and r*. Bernoulli, 4, 35-63.
  • [7] Bartlett, M.S. (1953a) Approximate confidence intervals. Biometrika, 40, 12-19.
  • [8] Bartlett, M.S. (1953b) Approximate confidence intervals. II. More than one unknown parameter. Biometrika, 40, 306-317.
  • [9] Bhattacharya, R.N. and Ghosh, J.K. (1978) On the validity of the formal Edgeworth expansion. Ann. Statist., 6, 434-451.
  • [10] Bickel, P.J., Chibisov, D.M. and van Zwet, W.R. (1981) On efficiency of first and second order. Internat. Statist. Rev., 49, 169-175.
  • [11] Chaganty, N.R. and Sethuraman, J. (1985) Large deviation local limit theorems for arbitrary sequences of random variables. Ann. Probab., 13, 97-114.
  • [12] Char, B.W., Geddes, K.O., Gonnet, G.H., Leong, B.L., Monagan, M.B. and Watt, S.M. (1991) Maple V. Language Reference Manual. Berlin: Springer-Verlag.
  • [13] Corcoran, S.A., Davison, A.C. and Spady, R.H. (1995) Reliable inference from empirical likelihoods. Preprint.
  • [14] Daniels, H.E. (1954) Saddlepoint approximations in statistics. Ann. Math. Statist., 25, 631-650.
  • [15] DiCiccio, T.J. and Romano, J.P. (1989) On adjustments based on the signed root of the empirical likelihood ratio statistic. Biometrika, 76, 447-456. Abstract can also be found in the ISI/STMA publication
  • [16] DiCiccio, T.J. and Stern, S.E. (1993) On Bartlett adjustments for approximate Bayesian inference. Biometrika, 80, 731-740. Abstract can also be found in the ISI/STMA publication
  • [17] DiCiccio, T.J. and Stern, S.E. (1994) Frequentist and Bayesian Bartlett correction of test statistics based on adjusted profile likelihoods. J. Roy. Statist. Soc. Ser. B, 56, 397-408. Abstract can also be found in the ISI/STMA publication
  • [18] DiCiccio, T.J., Hall P. and Romano, J.P. (1991) Empirical likelihood is Bartlett-correctable. Ann. Statist., 19, 1053-1061. Abstract can also be found in the ISI/STMA publication
  • [19] Goetze, F. and Hipp, C. (1983) Asymptotic expansions for sums of weakly dependent random vectors. Z. Wahrscheinlichkeitstheorie verw. Geb., 64, 211-239.
  • [20] Hall, P. (1992) The Bootstrap and Edgeworth Expansion. Berlin: Springer-Verlag.
  • [21] Hall, P. and Heyde, C.C. (1980) Martingale Limit Theory and its Application. New York: Academic Press.
  • [22] Jacod, J. and Shiryaev, A.N. (1987) Limit Theorems for Stochastic Processes. Berlin: Springer-Verlag.
  • [23] Jensen, J.L. (1992) The modified signed likelihood statistic and saddlepoint approximations. Biometrika, 79, 693-703. Abstract can also be found in the ISI/STMA publication
  • [24] Jensen, J.L. (1995) Saddlepoint Approximations in Statistics. Oxford: Oxford University Press.
  • [25] Jensen, J.L. (1997) A simple derivation of r* for curved exponential families. Scand. J. Statist., 24, 33-46. Abstract can also be found in the ISI/STMA publication
  • [26] Kong, A. and Cox, N.J. (1996) From efficient nonparametric tests for linkage analysis to semiparametric models and lodscores. Technical report no. 435, Department of Statistics, University of Chicago.
  • [27] Lawley, D.N. (1956) A general method for approximating the distribution of likelihood ratio criteria. Biometrika, 43, 295-303.
  • [28] Lazar, N. and Mykland, P.A. (1998) An evaluation of the power and conditionality properties of empirical likelihood. Biometrika, 85, 523-534. Abstract can also be found in the ISI/STMA publication
  • [29] McCullagh, P. (1984) Local sufficiency. Biometrika, 71, 233-244.
  • [30] McCullagh, P. (1987) Tensor Methods in Statistics. London: Chapman & Hall.
  • [31] McCullagh, P. and Nelder, J.A. (1989) Generalized Linear Models, 2nd edition. London: Chapman & Hall.
  • [32] McCullagh, P. and Tibshirani, R. (1990) A simple method for the adjustment of profile likelihoods. J. Roy. Statist. Soc. Ser. B, 52, 325-344. Abstract can also be found in the ISI/STMA publication
  • [33] McLeish, D.L. and Small, C.G. (1992) A projected likelihood function for semiparametric models. Biometrika, 79, 93-102. Abstract can also be found in the ISI/STMA publication
  • [34] Mykland, P.A. (1994) Bartlett type identities for martingales. Ann. Statist., 22, 21-38. Abstract can also be found in the ISI/STMA publication
  • [35] Mykland, P.A. (1995a) Dual likelihood. Ann. Statist., 23, 396-421. Abstract can also be found in the ISI/STMA publication
  • [36] Mykland, P.A. (1995b) Embedding and asymptotic expansions for martingales. Probab. Theory Related Fields, 103, 475-492. Abstract can also be found in the ISI/STMA publication
  • [37] Mykland, P.A. (1999) Bartlett identities and large deviations in likelihood theory. Ann. Statist., 27, 1105-1117. Abstract can also be found in the ISI/STMA publication
  • [38] Nicolae, D.L. (1999) Allele sharing models in gene mapping: a likelihood approach. Ph.D. thesis, University of Chicago.
  • [39] Peers, H.W. and Iqbal, M. (1985) Asymptotic expansions for confidence limits in the presence of nuisance parameters, with applications, J. Roy. Statist. Soc. Ser. B, 47, 547-554.
  • [40] Shenton, L.R. and Bowman, K.O. (1977) Maximum Likelihood Estimation in Small Samples. London: Griffin.
  • [41] Skovgaard, Ib (1986) A note on the differentiation of cumulants of log likelihood derivatives. Internat. Statist. Rev., 54, 29-32. Abstract can also be found in the ISI/STMA publication
  • [42] Skovgaard, Ib (1990) On the density of minimum contrast estimators. Ann. Statist., 18, 779-789. Abstract can also be found in the ISI/STMA publication
  • [43] Skovgaard, Ib (1996) An explicit large-deviation approximation to one-parameter tests. Bernoulli, 2, 145-165. Abstract can also be found in the ISI/STMA publication
  • [44] Stafford, J.E. (1994) Automating the partition of indexes. J. Comput. Graph. Statist., 3, 249-259. Abstract can also be found in the ISI/STMA publication
  • [45] Stafford, J.E. (1995) Exact cumulant calculations for Pearson ÷2 and Zelterman statistics for r-way contingency tables, J. Comput. Graph. Statist., 4, 199-212. Abstract can also be found in the ISI/STMA publication
  • [46] Stafford, J.E. and Andrews, D.F. (1993) A symbolic algorithm for studying adjustments to the profile likelihood. Biometrika, 80, 715-730. Abstract can also be found in the ISI/STMA publication
  • [47] Stafford, J.E., Andrews, D.F. and Wang. Y. (1994) Symbolic computation: a unified approach to studying likelihood. Statist. Comput., 4, 235-245. Abstract can also be found in the ISI/STMA publication
  • [48] Wallace, D.L. (1958) Asymptotic approximations to distributions. Ann. Math. Statist., 29, 635-654.