Optimal Tests for Combining p-Values
Source abstract
Abstract
Combining information (p-values) obtained from individual studies to test whether there is an overall effect is an important task in statistical data analysis. Many classical statistical tests, such as chi-square tests, can be viewed as being a p-value combination approach. It remains challenging to find powerful methods to combine p-values obtained from various sources. In this paper, we study a class of p-value combination methods based on gamma distribution. We show that this class of tests is optimal under certain conditions and several existing popular methods are equivalent to its special cases. An asymptotically and uniformly most powerful p-value combination test based on constrained likelihood ratio test is then studied. Numeric results from simulation study and real data examples demonstrate that the proposed tests are robust and powerful under many conditions. They have potential broad applications in statistical inference.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Advanced Statistical Methods and Models · Statistical Methods and Inference · Optimal Experimental Design Methods
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Department of Epidemiology and Biostatistics, School of Public Health, Indiana University Bloomington, 1025 E. 7th Street, Bloomington, IN 47405, USA
Funding
National Institutes of Health
References (43)
Show source-provided references
1. Fisher, R.A. (1932). Statistical Methods for Research Workers, Oliver and Boyd. [4th ed.].
2. On a New Method of Determining “Goodness of Fit”
3. Stouffer, S.A., Suchman, E.A., DeVinney, L.C., Star, S.A., and Williams, R.M. (1949). The American Soldier: Adjustment during Army Life. (Studies in Social Psychology in World War II), Princeton University Press.
4. Tippett, L.H.C. (1931). Methods of Statistics, Williams Norgate.
5. Is the weighted z-test the best method for combining probabilities from independent tests?10.1111/j.1420-9101.2010.02226.x ↗
6. A systematic comparison of methods for combining p-values from independent tests10.1016/j.csda.2003.11.020 ↗
7. Combining probability from independent tests: The weighted Z-method is superior to Fisher’s approach10.1111/j.1420-9101.2005.00917.x ↗
8. Cauchy combination test: A powerful test with analytic p-value calculation under arbitrary dependency structures10.1080/01621459.2018.1554485 ↗
9. Chen, Z. (2021). Robust tests for combining p-values under arbitrary dependency structures, unpublished.10.1038/s41598-022-07094-7 ↗
10. Karl Pearson’s meta-analysis revisited10.1214/09-AOS697 ↗
11. Hedges, L., and Olkin, I. (1985). Statistical Methods for Meta-Analysis, Academic.
12. Gene-based sequential burden association test10.1002/sim.8111 ↗
13. A novel gene-set association test based on variance-gamma distribution10.1177/0962280218791205 ↗
14. A genetic association test through combining two independent tests10.1016/j.ygeno.2018.07.010 ↗
15. Gene-based genetic association test with adaptive optimal weights10.1002/gepi.22098 ↗
16. A gene-based test of association through an orthogonal decomposition of genotype scores10.1007/s00439-017-1839-y ↗
17. Detecting associated single-nucleotide polymorphisms on the X chromosome in case control genome-wide association studies10.1177/0962280214551815 ↗
18. A powerful variant-set association test based on chi-square distribution10.1534/genetics.117.300287 ↗
19. Genetic association test based on principal component analysis10.1515/sagmb-2016-0061 ↗
20. Testing for gene-gene interaction in case-control GWAS10.4310/SII.2017.v10.n2.a10 ↗
21. A large multiethnic GWAS meta-analysis of cataract identifies new risk loci and sex-specific effects10.1038/s41467-021-23873-8 ↗
22. Genome-wide Association Study and Meta-analysis on Alcohol-Associated Liver Cirrhosis Identifies Genetic Risk Factors10.1002/hep.31535 ↗
23. Combining Independent Tests of Significance
24. Bonferroni, C. (1935). Il calcolo delle assicurazioni su gruppi di teste. Studi in Onore del Professore Salvatore Ortu Carboni, Bardi.
25. The combination of probabilities: An application of orthonormal functions10.1111/j.1467-842X.1961.tb00058.x ↗
26. On the optimally weighted z-test for combining probabilities from independent studies10.1016/j.csda.2013.09.005 ↗
27. Asymptotically optimal methods of combining tests10.1080/01621459.1979.10481035 ↗
28. Characterizations of complete classes of tests of some multiparametric hypotheses, with applications to likelihood ratio tests10.1214/aoms/1177728590 ↗
29. Rates of Convergence of Estimates and Test Statistics10.1214/aoms/1177698949 ↗
30. Asymptotic properties of maximum likelihood estimators and likelihood ratio tests under nonstandard conditions10.1080/01621459.1987.10478472 ↗
31. Inpatient rehabilitation specifically designed for geriatric patients: Systematic review and meta-analysis of randomised controlled trials10.1136/bmj.c1718 ↗
32. Interpretation of random effects meta-analyses10.1136/bmj.d549 ↗
33. Goodness-of-fit test for meta-analysis10.1038/srep16983 ↗
34. The Digitalis Investigation Group (1997). The effect of digoxin on mortality and morbidity in patients with heart failure. N. Engl. J. Med., 336, 525–533.10.1056/NEJM199702203360801 ↗
35. A two-stage procedure for comparing hazard rate functions10.1111/j.1467-9868.2007.00622.x ↗
36. Comparison of multiple hazard rate functions10.1111/biom.12412 ↗
37. Mosteller, F., Bush, R., and Lindzey, G. (1954). Handbook of Social Psychology, Addison-Wesley.
38. On the weighted combination of significance tests10.1111/j.2517-6161.1955.tb00201.x ↗
39. Van der Vaart, A.W. (2000). Asymptotic Statistics, Cambridge University Press.
40. Agresti, A. (2002). Categorical Data Analysis, Wiley-Interscience.10.1002/0471249688 ↗
41. The derivation and partition of χ2 in certain discrete distributions
42. Efficient simulation from a gamma distribution with small shape parameter10.1007/s00180-016-0692-0 ↗
43. Casella, G., and Berger, R.L. (2002). Statistical Inference, Duxbury.