Publication: Goodness-of-fit tests based on Rao's divergence under sparseness assumptions
Loading...
Full text at PDC
Publication Date
2002-08
Authors
Advisors (or tutors)
Editors
Journal Title
Journal ISSN
Volume Title
Publisher
Elsevier
Citations
Exportar
Abstract
In many practical situations the classical (fixed-cells) assumptions to test goodness-of-fit are inappropriate, and we consider an alternative set of assumptions, which we call sparseness assumptions. It is proved that, under general conditions, the proposed family of statistics based on Rao's divergence is asymptotically normal when the sample size n and the number of cells Mn tend to infinity so that n/Mn→ v > 0. This result is extended to contiguous alternatives, and subsequently it is possible to find the asymptotically most efficient member of the family.
Description
This work was supported by grant BMF 2000-0800.
UCM subjects
Unesco subjects
Keywords
Citation
M.S. Ali, D. Silvey, A general class of coefficients of divergence of one distribution from another,J.Roy. Statist. Soc. Ser. B 28 (1996) 131–140.
Y.M.M. Bishop, S.E. Fienberg, P.W. Holland, Discrete Multivariate Analysis, MIT Press, Cambridge, MA, 1975.
J. Burbea, C.R. Rao, On the convexity of some divergence measures based on entropy functions, IEEE Trans. Inform. Theory 28 (1982) 489–495.
J. Burbea, The Bose–Einstein entropy of degree a and its Jensen difference, Utilitas Math. 25 (1984) 225–240.
N. Cressie, T.R.C. Read, Multinomial goodness of fit test, J. Roy. Statist. Soc. Ser. B 46 (1984) 440–464.
I. Csiszar, Eine Informationtheoretische Ungleichung und ihre Anwendung auf den Bewis der Ergodizit€at von Markhoffschen Ketten, Publ. Math. Inst. Hungar. Acad. Sci. Ser. A 8 (1963) 85–108.
J. Hájek, Z. Sidák, Theory of Rank Tests, Academic Press, New York, 1967.
W. Hoeffding, Asymptotically optimal tests for multinomial distributions, Ann. Math. Statist. 36 (1965) 369–408.
L. Holst, Asymptotic normality and efficiency for certain goodness-of-fit tests, Biometrika 59 (1972) 137–145.
G.I. Ivchenko, Y.I. Medvedev, Separable statistics and hypothesis testing. The case of small samples, Theory Probab. Appl. 23 (1978) 764–765.
J.N. Kapur, Measures of uncertainty, mathematical programming and physics, J. Indian Soc. Agricultural Res. Statist. 24 (1972) 47–66.
L. LeCam, Locally Asymptotically Normal Families of Distributions, University of California Publications in Statistics, University of California Press, Berkeley, CA, 1960.
F. Liese, I. Vajda, ConvexStatistical Distances, Teubner (BSB), Leipzig, 1987.
M.L. Menendez, D. Morales, L. Pardo, I. Vajda, Asymptotic distributions of /-divergences of hypothetical and observed frequencies on refined partitions, Statist. Neerlandica 52 (1) (1998)71–89.
C. Morris, Admissible Bayes procedures and classes of epsilon Bayes procedures for testing hypothesis in a multinomial distribution, Technical Report 55, Department of Statistics, Stanford University, Stanford, CA, 1996.
C. Morris, Central limit theorems for multinomial sums, Ann. Statist. 3 (1975) 165–168.
M.C.Pardo,I.Vajda,About distances of discrete distributions satisfying the data proccesing theorem of information theory, Trans. IEEE Inform. Theory 43 (4) (1997) 1288–1293.
M.C. Pardo, On Burbea–Rao divergences based goodness-of-fit tests for multinomial model, J. Multivariate Anal. 69 (1999) 65–87.
T.R.C. Read, N.A.C. Cressie, Goodness of fit Statistics for Discrets Multivariate Data, Springer, Berlin, 1988.
I. Vajda, K. Vasek, Majorization, concave entropies, and comparison of experiments, Probab. Control Inform. Theory 14 (1985) 105–115.
K. Zografos, K. Ferentinos, T. Papaioannou, u-divergence statistics: sampling properties and multinomial goodness of fit divergence tests, Comm. Statist. Theory Methods 19 (5) (1990) 1785–1802.