Exact inference in contingency tables via stochastic approximation Monte Carlo

다국어 초록 (Multilingual Abstract)

Monte Carlo methods for the exact inference have received much attention recentlyin complete or incomplete contingency table analysis. However, conventional Markovchain Monte Carlo, such as the Metropolis–Hastings algorithm, and importance samplingmethods sometimes generate the poor performance by failing to produce valid tables. Inthis paper, we apply an adaptive Monte Carlo algorithm, the stochastic approximationMonte Carlo algorithm (SAMC; Liang, Liu, & Carroll, 2007), to the exact test of thegoodness-of-fit of the model in complete or incomplete contingency tables containingsome structural zero cells. The numerical results are in favor of our method in terms ofquality of estimates.

참고문헌 (Reference)

1 McCullagh, P., "The conditional distribution of goodness-of-fit statistics for discrete data" 81 : 104-107, 1986

2 Fienberg, S. E., "The analysis of cross-classified categorical data(2nd ed. )" Springer Science 2007

3 Liang, F., "Stochastic approximation in Monte Carlo computation" 102 (102): 305-320, 2007

4 Chen, H. F., "Stochastic approximation and its applications" Kluwer Academic Publishers 2002

5 Cheon, S., "Stochastic approximation Monte Carlo importance sampling for approximating exact conditional probabilities" Statistics and Computing 2013

6 Murdock, G. P., "Standard cross-cultural sample" 8 : 329-369, 1969

7 Andrieu, C., "Stability of stochastic approximation under verifiable conditions" 44 : 283-312, 2005

8 Rubinstein, R. Y., "Simulation and the Monte Carlo method(2nd ed. )" Wiley 2007

9 Chen, Y., "Sequential importance sampling for multiway tables" 34 : 523-545, 2006

10 Del Moral, P., "Sequential Monte Carlo samplers" 68 (68): 411-436, 2006

11 Chen, Y., "Sequential Monte Carlo methods for statistical analysis of tables" 100 : 109-120, 2005

12 Dinwoodie, I. H., "Sampling large tables with constraints" 21 : 1591-1609, 2011

13 Liang, F., "On the use of stochastic approximation Monte Carlo for Monte Carlo integration" 79 : 581-587, 2009

14 Hastings, W. K., "Monte Carlo sampling methods using Markov chains and their applications" 57 : 97-109, 1970

15 Aoki, S., "Markov chain Monte Carlo exact tests for incomplete two-way contingency tables" 75 (75): 787-812, 2005

16 Deloera, J. A., "Markov basis of three-way tables are arbitrarily complicated" 41 : 173-181, 2006

17 Dobra, A., "Markov bases for decomposable graphical models" 9 : 1093-1108, 2003

18 Rapallo, F., "Markov bases and subbases for bounded contingency tables" 62 (62): 785-805, 2010

19 Rapallo, F., "Markov bases and structural zeros" 41 : 164-172, 2006

20 Grizzle, J. E., "Loglinear models and tests of independence for contingency tables" 28 : 137-156, 1972

21 Hall, S. J. G., "Livestock breeds and their conservation : a global overview" 7 : 815-825, 1993

22 Bunea, F., "In Fields institute communications" American Mathematical Society 25-36, 2000

23 Whittaker, J., "Graphical models in applied mathematical statistics" Wiley 1990

24 Paul, S., "Goodness of fit of generalized linear models to sparse data" 62 : 323-333, 2000

25 Roberts, G. O., "Geometric convergence and central limit theorems for multidimensional Hastings and Metropolis algorithms" 83 : 95-110, 1996

26 Faraway, J. J., "Extending the linear model with R" Chapman & Hall/CRC 2006

27 Metropolis, N., "Equations of state calculations by fast computing machines" 21 : 1087-1091, 1953

28 Vidmar, N., "Effects of decision alternatives on the verdicts and social perceptions of simulated jurors" 22 : 211-218, 1972

29 Bishop, Y. M. M., "Discrete multivariate analysis : theory and practice" The MIT Press 1975

30 Chen, Y., "Conditional inference on tables with structural zeros" 16 : 445-467, 2007

31 Dobra, A., "Computing exact p-values in incomplete multi-way tables" Department of Statistics, University of Washington 2009

32 Agresti, A., "Categorical data analysis(2nd ed. )" Wiley 2002

33 Kreiner, S., "Analysis of multidimensional contingency tables by exact conditional tests : techniques and strategies" 14 : 97-112, 1987

34 Booth, J. G., "An importance sampling algorithm for exact conditional test in log-linear models" 86 (86): 321-332, 1999

35 White, D. R., "An exact significance test for three-way interaction effects" 18 (18): 103-122, 1983

36 Chen, Y., "Algebraic and geometric methods in statistics" The MIT Press 99-110, 2010

37 Diaconis, P., "Algebraic algorithms for sampling from conditional distributions" 26 : 363-397, 1998

38 Brunswick, A. F., "Adolescent health, sex, and fertility" 61 : 711-720, 1971

39 Haberman, S. J., "A warning on the use of chi-squared statistics with frequency tables with small expected cell counts" 83 : 555-560, 1988

40 Agresti, A., "A survey of exact inference for contingency tables" 7 : 131-153, 1992

41 Robbins, H., "A stochastic approximation method" 22 : 400-407, 1951

42 Caffo, B. S., "A Markov chain Monte Carlo algorithm for approximating exact conditional probabilities" 10 (10): 730-745, 2001

43 4ti2 team, "4ti2—a software package for algebraic, geometric and combinatorial problems on linear spaces"