Power evaluation of asymptotic tests for comparing two binomial proportions to detect direct and indirect association in large-scale studies

Abstract

Asymptotic tests are commonly used for comparing two binomial proportions when the sample size is sufficiently large. However, there is no consensus on the most powerful test. In this paper, we clarify this issue by comparing the power functions of three popular asymptotic tests: the Pearson’s χ² test, the likelihood-ratio test and the odds-ratio based test. Considering Taylor decompositions under local alternatives, the comparisons lead to recommendations on which test to use in view of both the experimental design and the nature of the investigated signal. We show that when the design is balanced between the two binomials, the three tests are equivalent in terms of power. However, when the design is unbalanced, differences in power can be substantial and the choice of the most powerful test also depends on the value of the parameters of the two compared binomials. We further investigated situations where the two binomials are not compared directly but through tag binomials. In these cases of indirect association, we show that the differences in power between the three tests are enhanced with decreasing values of the parameters of the tag binomials. Our results are illustrated in the context of genetic epidemiology where the analysis of genome-wide association studies provides insights regarding the low power for detecting rare variants.

Keywords

Test power genetic association studies likelihood ratio test odds ratio local alternatives

Get full access to this article

View all access options for this article.

References

Lydersen

Fagerland

Laake

. Recommended tests for association in 2 × 2 tables. Stat Med 2009; 28: 1159–1175.

Lin

Yang

. Improved p-value tests for comparing two independent binomial proportions. Commun Stat Simul Computat 2009; 38: 78–91.

Fagerland

Lydersen

Laake

. Recommended confidence intervals for two independent binomial proportions. Stat Meth Med Res 2015; 24(2): 224–254.

Hirji

Tau

Elashoff

. A quasi-exact test for comparing two binomial proportions. Stat Med 1991; 10: 1137–1153.

Agresti

. Categorical data analysis, 3rd ed. New York: Wiley, 2013.

Fisher

. On the interpretation of χ² from contingency tables, and the calculation of p. J R Stat Soc 1922; 85: 87–94.

Martín Andrés

Silva Mato

. Choosing the optimal unconditioned test for comparing two independent proportions. Computat Stat Data Anal 1994; 17: 555–574.

Berger

. Power comparison of exact unconditional tests for comparing two binomial proportions. Inst Stat Mimeo Ser 1994; 2266. . Technical report. Available at: http://www.stat.ncsu.edu/information/library/mimeo.php#2250.

Berger

. More powerful tests from confidence interval p-values. Am Stat 1996; 50: 314–318.

10.

Martín Andrés

Silva Mato

Tapia Garcí

et al.

Comparing the asymptotic power of exact tests in 2 × 2 tables. Computat Stat Data Anal 2004; 47: 745–756.

11.

Ruxton

Neuhäuser

. Good practice in testing for an association in contingency tables. Behavio Ecol Sociobiol 2010; 64: 1505–1513.

12.

Mehrotra

Chan

Berger

. A cautionary note on exact unconditional inference for a difference between two independent binomial proportions. Biometrics 2003; 59: 441–450.

13.

Cochran

. The chi-square test of goodness of fit. Ann Hum Genet 1952; 23: 315–345.

14.

Casella

Berger

. Statistical inference, 2nd ed. CA-Duxbury: Pacific Grove, 2002.

15.

Kang

Lee

. The sizes of the three popular asymptotic tests for testing homogeneity of two binomial proportions. Computat Stat Data Anal 2006; 51: 710–722.

16.

Agresti

Coull

. Approximate is better than “exact” for interval estimation of binomial proportions. Am Stat 1998; 52: 119–126.

17.

Cressie

Read

. Multinomial goodness-of-fit tests. J R Stat Soc B 1984; 46: 440–464.

18.

Cressie

Read

. Pearson’s χ² and the loglikelihood ratio statistic G²: a comparative review. Int Stat Rev 1989; 57: 19–43.

19.

Zar

. Biostatistical analysis, 5th ed. New York: Prentice Hall, 2008.

20.

Sokal

Rohlf

. Biometry, 1st ed. New York: Freeman, 1995.

21.

Quinn

Keough

. Experimental design and data analysis for biologists, 1st ed. Cambridge: Cambridge University Press, 2002.

22.

Drost

Kallenberg

Moore

et al.

Power approximations to multinomial tests of fit. J Am Stat Assoc 1989; 84: 130–141.

23.

Mitra

. On the limiting power function of the frequency chi-square test. Ann Math Stat 1958; 29: 1221–1233.

24.

Pickrell

Clerget-Darpoux

Bourgain

. Power of genome-wide association studies in the presence of interacting loci. Genet Epidemiol 2007; 31: 748–762.

25.

Carlson

Eberle

Rieder

et al.

Selecting a maximally informative set of single-nucleotide polymorphisms for association analyses using linkage disequilibrium. Am J Hum Genet 2004; 74: 106–120.

26.

Moskvina

O’Donovan

. Detailed analysis of the relative power of direct and indirect association studies and the implications for their interpretation. Hum Heredity 2007; 64: 63–73.

27.

WTCCC . Genome-wide association study of 14,000 cases of seven common diseases and 3,000 shared controls. Nature 2007; 447: 661–678.

28.

Zaykin

Kozbur

. P-value based analysis for shared controls design in genome-wide association studies. Genet Epidemiol 2010; 34: 725–738.

29.

Emily

. Indor: A new statistical procedure to test for snpxsnp epistasis in genome-wide association studies. Stat Med 2012; 31: 2359–2373.

30.

Kang

Ahn

. Tests for the homogeneity of two binomial proportions in extremely unbalanced 2 × 2 contingency tables. Stat Med 2008; 27: 2524–2535.

31.

Pritchard

Przeworski

. Linkage disequilibrium in humans: Models and data. Am J Hum Genet 2001; 69: 1–14.

32.

Purcell

Neale

Todd-Brown

et al.

Plink: a toolset for whole-genome association and population-based linkage analysis. Am J Hum Genet 2007; 81: 559–575.

33.

Franke

McGovern

DPB

Barrett

et al.

Genome-wide meta-analysis increases to 71 the number of confirmed crohn’s disease susceptibility loci. Nat Genet 2010; 42: 1118–1125.

34.

Benjamini

Hochberg

. Controlling the false discovery rate: A practical and powerful approach to multiple testing. J Royal Stat Soc Ser B (Methodological) 1995; 57: 289–300.

35.

Kang

Choi

Koh

. The effect of increasing control-to-case ratio on statistical power in a simulated case-control snp association study. Genom Inform 2009; 7: 148–151.

36.

Escott-Price

Ghodsi

Schmidt

. How allele frequency and study design affect association test statistics with misrepresentation errors. Biostatistics 2014; 15: 311–326.

37.

Bush WS and Moore JH. Genome-Wide association studies. PLoS Comput Biol 2012; 8(12): e1002822. DOI:10.1371/journal.pcbi.1002822.

38.

Soulakova

Roy

. On power approximations and comparison of several asymptotic tests to detect a specified difference between two proportions. Commun Stat – Simul Computat 2012; 41: 376–390.

39.

Guedj

Della-Chiesa

Picard

et al.

Computing power in case-control association studies through the use of quadratic approximations: Application to meta-statistics. Ann Hum Genet 2007; 71: 262–270.