Presanis, Anne M.; Ohlssen, David; Spiegelhalter, David J.; De Angelis, Daniela Conflict diagnostics in directed acyclic graphs, with applications in Bayesian evidence synthesis. (English) Zbl 1331.62160 Stat. Sci. 28, No. 3, 376-397 (2013). Summary: Complex stochastic models represented by directed acyclic graphs (DAGs) are increasingly employed to synthesise multiple, imperfect and disparate sources of evidence, to estimate quantities that are difficult to measure directly. The various data sources are dependent on shared parameters and hence have the potential to conflict with each other, as well as with the model. In a Bayesian framework, the model consists of three components: the prior distribution, the assumed form of the likelihood and structural assumptions. Any of these components may be incompatible with the observed data. The detection and quantification of such conflict and of data sources that are inconsistent with each other is therefore a crucial component of the model criticism process. We first review Bayesian model criticism, with a focus on conflict detection, before describing a general diagnostic for detecting and quantifying conflict between the evidence in different partitions of a DAG. The diagnostic is a \(p\)-value based on splitting the information contributing to inference about a “separator” node or group of nodes into two independent groups and testing whether the two groups result in the same inference about the separator node(s). We illustrate the method with three comprehensive examples: an evidence synthesis to estimate HIV prevalence; an evidence synthesis to estimate influenza case-severity; and a hierarchical growth model for rat weights. Cited in 14 Documents MSC: 62F15 Bayesian inference 05C05 Trees 62P10 Applications of statistics to biology and medical sciences; meta analysis Keywords:conflict; directed acyclic graph; evidence synthesis; graphical model; model criticism Software:BUGS; multcomp; BayesDA; R × Cite Format Result Cite Review PDF Full Text: DOI arXiv Euclid References: [1] Ades, A. E. and Cliffe, S. (2002). Markov chain Monte Carlo estimation of a multiparameter decision model: Consistency of evidence and the accurate assessment of uncertainty. Medical Decision Making 22 359-371. [2] Ades, A. E. and Sutton, A. J. (2006). Multiparameter evidence synthesis in epidemiology and medical decision-making: Current approaches. J. Roy. Statist. Soc. Ser. A 169 5-35. · doi:10.1111/j.1467-985X.2005.00377.x [3] Andrade, J. A. A. and O’Hagan, A. (2006). Bayesian robustness modeling using regularly varying distributions. Bayesian Anal. 1 169-188. · Zbl 1331.62175 · doi:10.1214/06-BA106 [4] Bayarri, M. J. and Berger, J. O. (1999). Quantifying surprise in the data and model verification. In Bayesian Statistics , 6 ( Alcoceber , 1998) (J. M. Bernardo, J. O. Berger, A. P. Dawid and A. F. M. Smith, eds.) 53-82. Oxford Univ. Press, New York. · Zbl 0974.62021 [5] Bayarri, M. J. and Berger, J. O. (2000). \(p\) values for composite null models. J. Amer. Statist. Assoc. 95 1127-1142, 1157-1170. · Zbl 1004.62022 · doi:10.2307/2669749 [6] Bayarri, M. J. and Castellanos, M. E. (2007). Bayesian checking of the second levels of hierarchical models. Statist. Sci. 22 322-343. · Zbl 1246.62029 · doi:10.1214/07-STS235 [7] Benjamini, Y. and Hochberg, Y. (1995). Controlling the false discovery rate: A practical and powerful approach to multiple testing. J. R. Stat. Soc. Ser. B Stat. Methodol. 57 289-300. · Zbl 0809.62014 [8] Benjamini, Y. and Yekutieli, D. (2001). The control of the false discovery rate in multiple testing under dependency. Ann. Statist. 29 1165-1188. · Zbl 1041.62061 · doi:10.1214/aos/1013699998 [9] Birrell, P. J., Ketsetzis, G., Gay, N. J., Cooper, B. S., Presanis, A. M., Harris, R. J., Charlett, A., Zhang, X.-S., White, P. J., Pebody, R. G. and De Angelis, D. (2011). Bayesian modeling to unmask and predict influenza A/H1N1pdm dynamics in London. Proc. Natl. Acad. Sci. USA 108 18238-18243. [10] Bousquet, N. (2008). Diagnostics of prior-data agreement in applied Bayesian analysis. J. Appl. Stat. 35 1011-1029. · Zbl 1474.62078 · doi:10.1080/02664760802192981 [11] Box, G. E. P. (1980). Sampling and Bayes’ inference in scientific modelling and robustness (with discussion). J. Roy. Statist. Soc. Ser. A 143 383-430. · Zbl 0471.62036 · doi:10.2307/2982063 [12] Box, G. E. P. and Tiao, G. C. (1992). Bayesian Inference in Statistical Analysis . Wiley, New York. · Zbl 0850.62004 [13] Bretz, F., Hothorn, T. and Westfall, P. (2011). Multiple Comparisons Using R , 1st ed. Chapman & Hall/CRC, London. · Zbl 1147.62355 [14] Clark, J. S., Bell, D., Chu, C., Courbaud, B., Dietze, M., Hersh, M., HilleRisLambers, J., Ibáñez, I., LaDeau, S., McMahon, S., Metcalf, J., Mohan, J., Moran, E., Pangle, L., Pearson, S., Salk, C., Shen, Z., Valle, D. and Wyckoff, P. (2010). High-dimensional coexistence based on individual variation: A synthesis of evidence. Ecological Monographs 80 569-608. [15] Cowell, R. G., Dawid, A. P., Lauritzen, S. L. and Spiegelhalter, D. J. (1999). Probabilistic Networks and Expert Systems . Springer, New York. · Zbl 0937.68121 · doi:10.1007/b97670 [16] Dahl, F. A., Gåsemyr, J. and Natvig, B. (2007). A robust conflict measure of inconsistencies in Bayesian hierarchical models. Scand. J. Stat. 34 816-828. · Zbl 1157.62011 [17] Dawid, A. P. (1984). Present position and potential developments: Some personal views: Statistical theory: The prequential approach. J. Roy. Statist. Soc. Ser. A 147 278-292. · Zbl 0557.62080 · doi:10.2307/2981683 [18] Dempster, A. P. (1997). The direct use of likelihood for significance testing. Statist. Comput. 7 247-252. [19] Dias, S., Welton, N. J., Caldwell, D. M. and Ades, A. E. (2010). Checking consistency in mixed treatment comparison meta-analysis. Stat. Med. 29 932-944. · doi:10.1002/sim.3767 [20] DuMouchel, W. H. and Harris, J. E. (1983). Bayes methods for combining the results of cancer studies in humans and other species. J. Amer. Statist. Assoc. 78 293-315. · Zbl 0528.62089 · doi:10.2307/2288631 [21] Evans, M. (1997). Bayesian inference procedures derived via the concept of relative surprise. Comm. Statist. Theory Methods 26 1125-1143. · Zbl 0934.62026 · doi:10.1080/03610929708831972 [22] Evans, M. and Jang, G. H. (2010). Invariant \(P\)-values for model checking. Ann. Statist. 38 512-525. · Zbl 1181.62030 · doi:10.1214/09-AOS727 [23] Evans, M. and Jang, G. H. (2011). Weak informativity and the information in one prior relative to another. Statist. Sci. 26 423-439. · Zbl 1246.62007 · doi:10.1214/11-STS357 [24] Evans, M. and Moshonov, H. (2006). Checking for prior-data conflict. Bayesian Anal. 1 893-914 (electronic). · Zbl 1331.62030 · doi:10.1214/06-BA129 [25] Evans, M. and Moshonov, H. (2007). Checking for prior-data conflict with hierarchically specified priors. In Bayesian Statistics and Its Applications (A. K. Upadhyay, U. Singh and D. Dey, eds.) 145-159. Anamaya Publishers, New Delhi. · Zbl 1331.62030 · doi:10.1214/06-BA129 [26] Gåsemyr, J. and Natvig, B. (2009). Extensions of a conflict measure of inconsistencies in Bayesian hierarchical models. Scand. J. Stat. 36 822-838. · Zbl 1222.62037 · doi:10.1111/j.1467-9469.2009.00659.x [27] Gamerman, D. and Lopes, H. F. (2006). Markov Chain Monte Carlo : Stochastic Simulation for Bayesian Inference , 2nd ed. Chapman & Hall/CRC, Boca Raton, FL. · Zbl 1137.62011 [28] Gelfand, A. E., Hills, S. E., Racine-Poon, A. and Smith, A. F. M. (1990). Illustration of Bayesian inference in normal data models using Gibbs sampling. J. Amer. Statist. Assoc. 85 972-985. [29] Gelman, A., Meng, X.-L. and Stern, H. (1996). Posterior predictive assessment of model fitness via realized discrepancies. Statist. Sinica 6 733-807. · Zbl 0859.62028 [30] Gelman, A., Carlin, J. B., Stern, H. S. and Rubin, D. B. (2003). Bayesian Data Analysis , 2nd ed. Chapman & Hall/CRC, London. · Zbl 1279.62004 [31] Gneiting, T. and Raftery, A. E. (2007). Strictly proper scoring rules, prediction, and estimation. J. Amer. Statist. Assoc. 102 359-378. · Zbl 1284.62093 · doi:10.1198/016214506000001437 [32] Greenland, S. (2009). Relaxation penalties and priors for plausible modeling of nonidentified bias sources. Statist. Sci. 24 195-210. · Zbl 1328.62051 · doi:10.1214/09-STS291 [33] Henderson, D. A., Boys, R. J. and Wilkinson, D. J. (2010). Bayesian calibration of a stochastic kinetic computer model using multiple data sources. Biometrics 66 249-256. · Zbl 1187.62054 · doi:10.1111/j.1541-0420.2009.01245.x [34] Higgins, J. P. T., Jackson, D., Barrett, J. K., Lu, G., Ades, A. E. and White, I. R. (2012). Consistency and inconsistency in network meta-analysis: Concepts and models for multi-arm studies. Research Synthesis Methods 3 98-110. [35] Hjort, N. L., Dahl, F. A. and Steinbakk, G. H. (2006). Post-processing posterior predictive \(p\)-values. J. Amer. Statist. Assoc. 101 1157-1174. · Zbl 1120.62307 · doi:10.1198/016214505000001393 [36] Hothorn, T., Bretz, F. and Westfall, P. (2008). Simultaneous inference in general parametric models. Biom. J. 50 346-363. · Zbl 1147.62355 · doi:10.1002/bimj.200810425 [37] Jackson, C., Richardson, S. and Best, N. (2008). Studying place effects on health by synthesising individual and area-level outcomes. Social Science & Medicine 67 1995-2006. [38] Jackson, D., White, I. R. and Carpenter, J. (2012). Identifying influential observations in Bayesian models by using Markov chain Monte Carlo. Stat. Med. 31 1238-1248. · doi:10.1002/sim.4356 [39] Johnson, V. E. (2007). Bayesian model assessment using pivotal quantities. Bayesian Anal. 2 719-733. · Zbl 1331.62147 · doi:10.1214/07-BA229 [40] Jones, H. E., Ohlssen, D. I. and Spiegelhalter, D. J. (2008). Use of the false discovery rate when comparing multiple health care providers. J. Clin. Epidemiol. 61 232-240. [41] Kass, R. E. (1990). Data-translated likelihood and Jeffreys’s rules. Biometrika 77 107-114. · Zbl 0692.62001 [42] Langford, I. H. and Lewis, T. (1998). Outliers in multilevel data. J. Roy. Statist. Soc. Ser. A 161 121-160. [43] Lauritzen, S. L. (1996). Graphical Models. Oxford Statistical Science Series 17 . Oxford Univ. Press, New York. · Zbl 0907.62001 [44] Lu, G. and Ades, A. E. (2006). Assessing evidence inconsistency in mixed treatment comparisons. J. Amer. Statist. Assoc. 101 447-459. · Zbl 1119.62354 · doi:10.1198/016214505000001302 [45] Lunn, D., Spiegelhalter, D., Thomas, A. and Best, N. (2009). The BUGS project: Evolution, critique and future directions. Stat. Med. 28 3049-3067. · doi:10.1002/sim.3680 [46] Marshall, E. C. and Spiegelhalter, D. J. (2007). Identifying outliers in Bayesian hierarchical models: A simulation-based approach. Bayesian Anal. 2 409-444. · Zbl 1331.62032 · doi:10.1214/07-BA218 [47] O’Hagan, A. (2003). HSSS model criticism (with discussion). In Highly Structured Stochastic Systems , 1st ed. (P. J. Green, N. L. Hjort and S. Richardson, eds.). Oxford Univ. Press, New York. [48] Ohlssen, D. I., Sharples, L. D. and Spiegelhalter, D. J. (2007). A hierarchical modelling framework for identifying unusual performance in health care providers. J. Roy. Statist. Soc. Ser. A 170 865-890. · doi:10.1111/j.1467-985X.2007.00487.x [49] Presanis, A. M., De Angelis, D., Spiegelhalter, D. J., Seaman, S., Goubar, A. and Ades, A. E. (2008). Conflicting evidence in a Bayesian synthesis of surveillance data to estimate human immunodeficiency virus prevalence. J. Roy. Statist. Soc. Ser. A 171 915-937. · doi:10.1111/j.1467-985X.2008.00543.x [50] Presanis, A. M., Pebody, R. G., Paterson, B. J., Tom, B. D. M., Birrell, P. J., Charlett, A., Lipsitch, M. and De Angelis, D. (2011). Changes in severity of 2009 pandemic A/H1N1 influenza in England: A Bayesian evidence synthesis. BMJ 343 d5408+. [51] R Development Core Team. (2005). R : A Language and Environment for Statistical Computing . Vienna, Austria. [52] Robins, J. M., van der Vaart, A. and Ventura, V. (2000). Asymptotic distribution of \(p\) values in composite null models. J. Amer. Statist. Assoc. 95 1143-1167, 1171-1172. · Zbl 1072.62522 · doi:10.2307/2669750 [53] Rubin, D. B. (1984). Bayesianly justifiable and relevant frequency calculations for the applied statistician. Ann. Statist. 12 1151-1172. · Zbl 0555.62010 · doi:10.1214/aos/1176346785 [54] Scheel, I., Green, P. J. and Rougier, J. C. (2011). A graphical diagnostic for identifying influential model choices in Bayesian hierarchical models. Scand. J. Stat. 38 529-550. · Zbl 1246.62064 · doi:10.1111/j.1467-9469.2010.00717.x [55] Spiegelhalter, D. J., Abrams, K. R. and Myles, J. P. (2004). Bayesian Approaches to Clinical Trials and Health-Care Evaluation . Wiley, New York. [56] Spiegelhalter, D. J., Dawid, A. P., Lauritzen, S. L. and Cowell, R. G. (1993). Bayesian analysis in expert systems. Statist. Sci. 8 219-283. · Zbl 0955.62523 · doi:10.1214/ss/1177010888 [57] Spiegelhalter, D. J., Harris, N. L., Bull, K. and Franklin, R. C. G. (1994). Empirical evaluation of prior beliefs about frequencies: Methodology and a case study in congenital heart disease. J. Amer. Statist. Assoc. 89 435-443. [58] Spiegelhalter, D. J., Best, N. G., Carlin, B. P. and van der Linde, A. (2002). Bayesian measures of model complexity and fit. J. R. Stat. Soc. Ser. B Stat. Methodol. 64 583-639. · Zbl 1067.62010 · doi:10.1111/1467-9868.00353 [59] Steinbakk, G. H. andStorvik, G. O. (2009). Posterior predictive \(p\)-values in Bayesian hierarchical models. Scand. J. Stat. 36 320-336. · Zbl 1190.62061 · doi:10.1111/j.1467-9469.2008.00630.x [60] Turner, R. M., Spiegelhalter, D. J., Smith, G. C. S. and Thompson, S. G. (2009). Bias modelling in evidence synthesis. J. Roy. Statist. Soc. Ser. A 172 21-47. · doi:10.1111/j.1467-985X.2008.00547.x [61] Welton, N. J., Ades, A. E., Carlin, J. B., Altman, D. G. and Sterne, J. A. C. (2009). Models for potentially based evidence in meta-analysis using empirically based priors. J. Roy. Statist. Soc. Ser. A 172 119-136. · doi:10.1111/j.1467-985X.2008.00548.x [62] Welton, N. J., Sutton, A. J., Cooper, N. J., Abrams, K. R. and Ades, A. E. (2012). Evidence Synthesis in a Decision Modelling Framework. In Evidence Synthesis for Decision Making in Healthcare 138-150. Wiley, New York. [63] White, I. R., Barrett, J. K., Jackson, D. and Higgins, J. P. T. (2012). Consistency and inconsistency in network meta-analysis: Model estimation using multivariate meta-regression. Research Synthesis Methods 3 111-125. This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.