• Bernoulli
  • Volume 16, Number 4 (2010), 1208-1223.

Second order ancillary: A differential view from continuity

Ailana M. Fraser, D.A.S. Fraser, and Ana-Maria Staicu

Full-text: Open access


Second order approximate ancillaries have evolved as the primary ingredient for recent likelihood development in statistical inference. This uses quantile functions rather than the equivalent distribution functions, and the intrinsic ancillary contour is given explicitly as the plug-in estimate of the vector quantile function. The derivation uses a Taylor expansion of the full quantile function, and the linear term gives a tangent to the observed ancillary contour. For the scalar parameter case, there is a vector field that integrates to give the ancillary contours, but for the vector case, there are multiple vector fields and the Frobenius conditions for mutual consistency may not hold. We demonstrate, however, that the conditions hold in a restricted way and that this verifies the second order ancillary contours in moderate deviations. The methodology can generate an appropriate exact ancillary when such exists or an approximate ancillary for the numerical or Monte Carlo calculation of $p$-values and confidence quantiles. Examples are given, including nonlinear regression and several enigmatic examples from the literature.

Article information

Bernoulli, Volume 16, Number 4 (2010), 1208-1223.

First available in Project Euclid: 18 November 2010

Permanent link to this document

Digital Object Identifier

Mathematical Reviews number (MathSciNet)

Zentralblatt MATH identifier

approximate ancillary approximate location model conditioning confidence $p$-value quantile


Fraser, Ailana M.; Fraser, D.A.S.; Staicu, Ana-Maria. Second order ancillary: A differential view from continuity. Bernoulli 16 (2010), no. 4, 1208--1223. doi:10.3150/10-BEJ248.

Export citation


  • [1] Andrews, D.F., Fraser, D.A.S. and Wong, A. (2005). Computation of distribution functions from likelihood information near observed data. J. Statist. Plann. Inference 134 180–193.
  • [2] Barndorff-Nielsen, O.E. (1986). Inference on full or partial parameters based on the standardized log likelihood ratio. Biometrika 73 307–322.
  • [3] Barndorff-Nielsen, O.E. (1987). Discussion of “Parameter orthogonality and approximate conditional inference.” J. R. Stat. Soc. Ser. B Stat. Methodol. 49 18–20.
  • [4] Berger, J.O. and Sun, D. (2008). Objective priors for the bivariate normal model. Ann. Statist. 36 963–982.
  • [5] Cakmak, S., Fraser, D.A.S. and Reid, N. (1994). Multivariate asymptotic model: Exponential and location approximations. Util. Math. 46 21–31.
  • [6] Cheah, P.K., Fraser, D.A.S. and Reid, N. (1995). Adjustment to likelihood and densities: Calculating significance. J. Statist. Res. 29 1–13.
  • [7] Cox, D.R. (1980). Local ancillarity. Biometrika 67 279–286.
  • [8] Cox, D.R. and Reid, N. (1987). Parameter orthogonality and approximate conditional inference. J. R. Stat. Soc. Ser. B Stat. Methodol. 49 1–39.
  • [9] Daniels, H.E. (1954). Saddle point approximations in statistics. Ann. Math. Statist. 25 631–650.
  • [10] Fisher, R.A. (1925). Theory of statistical estimation. Proc. Camb. Phil. Soc. 22 700–725.
  • [11] Fisher, R.A. (1934). Two new properties of mathematical likelihood. Proc. R. Soc. Lond. Ser. A 144 285–307.
  • [12] Fisher, R.A. (1935). The logic of inductive inference. J. R. Stat. Soc. Ser. B Stat. Methodol. 98 39–54.
  • [13] Fisher, R.A. (1956). Statistical Methods and Scientific Inference. Edinburgh: Oliver & Boyd.
  • [14] Fraser, D.A.S. (1979). Inference and Linear Models. New York: McGraw-Hill.
  • [15] Fraser, D.A.S. (1993). Directional tests and statistical frames. Statist. Papers 34 213–236.
  • [16] Fraser, D.A.S. (2003). Likelihood for component parameters. Biometrika 90 327–339.
  • [17] Fraser, D.A.S. (2004). Ancillaries and conditional inference, with discussion. Statist. Sci. 19 333–369.
  • [18] Fraser, D.A.S. and Reid, N. (1995). Ancillaries and third order significance. Util. Math. 47 33–53.
  • [19] Fraser, D.A.S. and Reid, N. (2001). Ancillary information for statistical inference. In Empirical Bayes and Likelihood Inference (S.E. Ahmed and N. Reid, eds.) 185–207. New York: Springer.
  • [20] Fraser, D.A.S. and Reid, N. (2002). Strong matching for frequentist and Bayesian inference. J. Statist. Plann. Inference 103 263–285.
  • [21] Fraser, D.A.S., Reid, N., Marras, E. and Yi, G.Y. (2010). Default priors for Bayes and frequentist inference. J. R. Stat. Soc. Ser. B Stat. Methodol. To appear.
  • [22] Fraser, D.A.S., Reid, N. and Wu, J. (1999). A simple general formula for tail probabilities for Bayes and frequentist inference. Biometrika 86 249–264.
  • [23] Fraser, D.A.S. and Rousseau, J. (2008). Studentization and deriving accurate p-values. Biometrika 95 1–16.
  • [24] Fraser, D.A.S., Wong, A. and Wu, J. (1999). Regression analysis, nonlinear or nonnormal: Simple and accurate p-values from likelihood analysis. J. Amer. Statist. Assoc. 94 1286–1295.
  • [25] Lugannani, R. and Rice, S. (1980). Saddlepoint approximation for the distribution of the sum of independent random variables. Adv. in Appl. Probab. 12 475–490.
  • [26] McCullagh, P. (1984). Local sufficiency. Biometrika 71 233–244.
  • [27] McCullagh, P. (1992). Conditional inference and Cauchy models. Biometrika 79 247–259.
  • [28] Reid, N. and Fraser, D.A.S. (2010). Mean likelihood and higher order inference. Biometrika 97. To appear.
  • [29] Severini, T.A. (2001). Likelihood Methods in Statistics. Oxford: Oxford Univ. Press.