## Electronic Journal of Statistics

### Gradient statistic: Higher-order asymptotics and Bartlett-type correction

#### Abstract

We obtain an asymptotic expansion for the null distribution function of the gradient statistic for testing composite null hypotheses in the presence of nuisance parameters. The expansion is derived using a Bayesian route based on the shrinkage argument described in [10]. Using this expansion, we propose a Bartlett-type corrected gradient statistic with chi-square distribution up to an error of order $o(n^{-1})$ under the null hypothesis. Further, we also use the expansion to modify the percentage points of the large sample reference chi-square distribution. Monte Carlo simulation experiments and various examples are presented and discussed.

#### Article information

Source
Electron. J. Statist., Volume 7 (2013), 43-61.

Dates
First available in Project Euclid: 11 January 2013

https://projecteuclid.org/euclid.ejs/1357913281

Digital Object Identifier
doi:10.1214/12-EJS763

Mathematical Reviews number (MathSciNet)
MR3020413

Zentralblatt MATH identifier
1336.62143

#### Citation

Vargas, Tiago M.; Ferrari, Silvia L.P.; Lemonte, Artur J. Gradient statistic: Higher-order asymptotics and Bartlett-type correction. Electron. J. Statist. 7 (2013), 43--61. doi:10.1214/12-EJS763. https://projecteuclid.org/euclid.ejs/1357913281

#### References

• [1] Bai, P. (2009). Sphericity test in a GMANOVA-MANOVA model with normal error., Journal of Multivariate Analysis 100, 2305–2312.
• [2] Bickel, P.J., Ghosh, J.K. (1990). A decomposition for the likelihood ratio statistic and the Bartlett correction - a Bayesian argument., Annals of Statistics 18, 1070–1090.
• [3] Birnbaum, Z.W., Saunders, S.C. (1969). A new family of life distributions., Journal of Applied Probability 6, 319–327.
• [4] Chang, H.I., Mukerjee, R. (2010). Highest posterior density regions with approximate frequentist validity: the role of data-dependent priors., Statistics and Probability Letters 80, 1791–1797.
• [5] Chang, H.I., Mukerjee, R. (2011). Data-dependent probability matching priors for likelihood ratio and adjusted likelihood ratio statistics., Statistics. In press, DOI:10.1080/02331888.2011.587880.
• [6] Cordeiro, G.M., Ferrari, S.L.P. (1991). A modified score test statistic having chi-squared distribuition to order $n^-1$., Biometrika 78, 573–582.
• [7] Cordeiro, G.M., Cribari-Neto, F. (1996). On Bartlett and Bartlett-type corrections., Econometric Reviews 15, 339–367.
• [8] Cox, D.R., Reid, N. (1987). Parameter orthogonality and approximate conditional inference (with discussion)., Journal of the Royal Statistical Society B 40, 1–39.
• [9] Datta, G.S., Mukerjee, R. (2003)., Probability Matching Priors: Higher Order Asymptoptics. Springer-Verlag: New York.
• [10] Ghosh, J.K., Mukerjee, R. (1991). Characterization of priors under wich Bayesian and frequentist Bartlett corrections are equivalent in the multiparameter case., Journal of Multivariate Analysis 38, 385–393.
• [11] Harris, P. (1985). An asymptotic expansion for the null distribution of the efficient score statistic., Biometrika 72, 653–659.
• [12] Hayakawa, T. (1977). The likelihood ratio criterion and the asymptotic expansion of its distribution., Annals of the Institute of Statistical Mathematics 29, 359–378.
• [13] Hill, G.W., Davis, A.W. (1968). Generalized asymptotic expansions of Cornish–Fisher type., The Annals of Mathematical Statistics 39, 1264-73.
• [14] Lagos, B.M., Morettin, P.A. (2004). Improvement of the likelihood ratio test statistic in ARMA models., Journal of Time Series Analysis 25, 83–101.
• [15] Lagos, B.M., Morettin, P.A., Barroso, L.P. (2010). Some corrections of the score test statistic for gaussian ARMA models., Brazilian Journal of Probability and Statistics 24, 434–456.
• [16] Lawley, D. (1956). A general method for approximating to the distribution of likelihood ratio criteria., Biometrika 43, 295–303.
• [17] Lemonte, A.J. (2011). Local power of some tests in exponential family nonlinear models., Journal of Statistical Planning and Inference 141, 1981–1989.
• [18] Lemonte, A.J. (2012). Local power properties of some asymptotic tests in symmetric linear regression models., Journal of Statistical Planning and Inference 142, 1178–1188.
• [19] Lemonte, A.J., Ferrari, S.L.P. (2012a). The local power of the gradient test., Annals of the Institute of Statistical Mathematics 64, 373–381.
• [20] Lemonte, A.J., Ferrari, S.L.P. (2012b). A note on the local power of the LR, Wald, score and gradient tests., Electronic Journal of Statistics 6, 421–434.
• [21] Mukerjee, R., Reid, N. (2000). On the Bayesian approach for frequentist computations., Brazilian Journal of Probability and Statistics 14, 159–166.
• [22] Noma, H. (2011). Confidence intervals for a random-effects meta-analysis based on Bartlett-type corrections., Statistics in Medicine 30, 3304–3312.
• [23] Rao, C.R. (1948). Large sample tests of statistical hypotheses concerning several parameters with applications to problens of estimation., Proceedings of the Cambridge Philosophical Society 44, 50–57.
• [24] Rao, C.R. (2005). Score test: historical review and recent developments. In, Advances in Ranking and Selection, Multiple Comparisons, and Reliability, N. Balakrishnan, N. Kannan and H. N. Nagaraja, eds. Birkhuser, Boston.
• [25] Terrell, G.R. (2002). The gradient statistic., Computing Science and Statistics 34, 206–215.
• [26] Tu, D., Chen, J., Shi, P., Wu, Y. (2005). A Bartlett type correction for Rao’s score test in Cox regression model., Sankhya 67, 722–735.
• [27] van Giersbergen, N.P.A. (2009). Bartlett correction in the stable AR(1) model with intercept and trend., Econometric Theory 25, 857–872.
• [28] Wald, A. (1943). Tests of statistical hypothesis concerning several parameters when the number of observations is large., Transactions of the American Mathematical Society 54, 426–482.
• [29] Wilks, S.S. (1938). The large-sample distribution of the likelihood ratio for testing composite hypothesis., Annals of Mathematical Statistics 9, 60–62.
• [30] Zucker, D.M., Lieberman, O., Manor, O. (2000). Improved small sample inference in the mixed linear model: Bartlett correction and adjusted likelihood., Journal of the Royal Statistical Society B 62, 827–838.