×

Robust Bayesian synthetic likelihood via a semi-parametric approach. (English) Zbl 1436.62090

Summary: Bayesian synthetic likelihood (BSL) is now a well-established method for performing approximate Bayesian parameter estimation for simulation-based models that do not possess a tractable likelihood function. BSL approximates an intractable likelihood function of a carefully chosen summary statistic at a parameter value with a multivariate normal distribution. The mean and covariance matrix of this normal distribution are estimated from independent simulations of the model. Due to the parametric assumption implicit in BSL, it can be preferred to its nonparametric competitor, approximate Bayesian computation, in certain applications where a high-dimensional summary statistic is of interest. However, despite several successful applications of BSL, its widespread use in scientific fields may be hindered by the strong normality assumption. In this paper, we develop a semi-parametric approach to relax this assumption to an extent and maintain the computational advantages of BSL without any additional tuning. We test our new method, semiBSL, on several challenging examples involving simulated and real data and demonstrate that semiBSL can be significantly more robust than BSL and another approach in the literature.

MSC:

62F15 Bayesian inference
62F35 Robustness and adaptive procedures (parametric inference)
62H05 Characterization and structure theory for multivariate probability distributions; copulas

Software:

ABC-SubSim; glasso
PDFBibTeX XMLCite
Full Text: DOI arXiv Link

References:

[1] An, Z.; South, Lf; Nott, Dj; Drovandi, Cc, Accelerating Bayesian synthetic likelihood with the graphical lasso, J. Comput. Graph. Stat., 28, 2, 471-475 (2019) · Zbl 07499068 · doi:10.1080/10618600.2018.1537928
[2] Anderson, C.; Coles, S., The largest inclusions in a piece of steel, Extremes, 5, 3, 237-252 (2002) · Zbl 1036.62115 · doi:10.1023/A:1024025027522
[3] Andrieu, C.; Roberts, Go, The pseudo-marginal approach for efficient Monte Carlo computations, Ann. Stat., 37, 2, 697-725 (2009) · Zbl 1185.60083 · doi:10.1214/07-AOS574
[4] Barbu, Cm; Sethuraman, K.; Billig, Emw; Levy, Mz, Two-scale dispersal estimation for biological invasions via synthetic likelihood, Ecography, 41, 4, 661-672 (2017) · doi:10.1111/ecog.02575
[5] Bedford, T.; Cooke, Rm, Vines-a new graphical model for dependent random variables, Ann. Stat., 30, 4, 1031-1068 (2002) · Zbl 1101.62339 · doi:10.1214/aos/1031689016
[6] Blum, Mgb; François, O., Non-linear regression models for approximate Bayesian computation, Stat. Comput., 20, 1, 63-73 (2010) · doi:10.1007/s11222-009-9116-0
[7] Bortot, P.; Coles, Sg; Sisson, Sa, Inference for stereological extremes, J. Am. Stat. Assoc., 102, 477, 84-92 (2007) · Zbl 1284.62795 · doi:10.1198/016214506000000988
[8] Boudt, K.; Cornelissen, J.; Croux, C., The Gaussian rank correlation estimator: robustness properties, Stat. Comput., 22, 2, 471-483 (2012) · Zbl 1322.62129 · doi:10.1007/s11222-011-9237-0
[9] Chambers, Jm; Mallows, Cl; Stuck, Bw, A method for simulating stable random variables, J. Am. Stat. Assoc., 71, 354, 340-344 (1976) · Zbl 0341.65003 · doi:10.1080/01621459.1976.10480344
[10] Chiachio, M.; Beck, J.; Chiachio, J.; Rus, G., Approximate Bayesian computation by subset simulation, SIAM J. Sci. Comput., 36, 3, A1339-A1358 (2014) · Zbl 1297.65013 · doi:10.1137/130932831
[11] Drovandi, Cc; Pettitt, An; Faddy, Mj, Approximate Bayesian computation using indirect inference, J. R. Stat. Soc. Ser. C (Appl. Stat.), 60, 3, 317-337 (2011) · doi:10.1111/j.1467-9876.2010.00747.x
[12] Drovandi, Cc; Pettitt, An; Lee, A., Bayesian indirect inference using a parametric auxiliary model, Stat. Sci., 30, 1, 72-95 (2015) · Zbl 1332.62088 · doi:10.1214/14-STS498
[13] Dutta, R., Corander, J., Kaski, S., Gutmann, M.U.: Likelihood-free inference by ratio estimation. ArXiv preprint arXiv:1611.10242v3 (2017) · Zbl 1384.62089
[14] Epanechnikov, Va, Non-parametric estimation of a multivariate probability density, Theory Probab. Appl., 14, 1, 153-158 (1969) · doi:10.1137/1114019
[15] Everitt, RG.: Bootstrapped synthetic likelihood. ArXiv preprint arXiv:1711.05825v2 (2017)
[16] Fasiolo, M.; Wood, Sn; Hartig, F.; Bravington, Mv, An extended empirical saddlepoint approximation for intractable likelihoods, Electron. J. Stat., 12, 1, 1544-1578 (2018) · Zbl 1395.62338 · doi:10.1214/18-EJS1433
[17] Fearnhead, P.; Prangle, D., Constructing summary statistics for approximate Bayesian computation: semi-automatic approximate Bayesian computation, J. R. Stat. Soc. Ser. B (Stat. Methodol.), 74, 3, 419-474 (2012) · Zbl 1411.62057 · doi:10.1111/j.1467-9868.2011.01010.x
[18] Friedman, J.; Hastie, T.; Tibshirani, R., Sparse inverse covariance estimation with the graphical lasso, Biostatistics, 9, 3, 432-441 (2008) · Zbl 1143.62076 · doi:10.1093/biostatistics/kxm045
[19] Gleim, A., Pigorsch, C.: Approximate Bayesian computation with indirect summary statistics. Draft paper: http://www.ect-pigorschmeeuni-bonnde/data/research/papers (2013)
[20] Jones, Mc; Pewsey, A., Sinh-arcsinh distributions, Biometrika, 96, 4, 761-780 (2009) · Zbl 1183.62019 · doi:10.1093/biomet/asp053
[21] Li, J.; Nott, D.; Fan, Y.; Sisson, S., Extending approximate Bayesian computation methods to high dimensions via a Gaussian copula model, Comput. Stat. Data Anal., 106, 77-89 (2017) · Zbl 1466.62136 · doi:10.1016/j.csda.2016.07.005
[22] Liu, H.; Lafferty, J.; Wasserman, L., The nonparanormal: semiparametric estimation of high dimensional undirected graphs, J. Mach. Learn. Res., 10, 2295-2328 (2009) · Zbl 1235.62035
[23] Marchand, P.; Boenke, M.; Green, Dm, A stochastic movement model reproduces patterns of site fidelity and long-distance dispersal in a population of Fowler’s toads (Anaxyrus fowleri), Ecol. Model., 360, 63-69 (2017) · doi:10.1016/j.ecolmodel.2017.06.025
[24] Ong, Vmh; Nott, Dj; Tran, Mn; Sisson, Sa; Drovandi, Cc, Likelihood-free inference in high dimensions with synthetic likelihood, Comput. Stat. Data Anal., 128, 271-291 (2018) · Zbl 1469.62123 · doi:10.1016/j.csda.2018.07.008
[25] Ong, Vmh; Nott, Dj; Tran, Mn; Sisson, Sa; Drovandi, Cc, Variational Bayes with synthetic likelihood, Stat. Comput., 28, 4, 971-988 (2018) · Zbl 1384.65015 · doi:10.1007/s11222-017-9773-3
[26] Parzen, E., On estimation of a probability density function and mode, Ann. Math. Stat., 33, 3, 1065-1076 (1962) · Zbl 0116.11302 · doi:10.1214/aoms/1177704472
[27] Price, Lf; Drovandi, Cc; Lee, A.; Nott, Dj, Bayesian synthetic likelihood, J. Comput. Graph. Stat., 27, 1-11 (2018) · Zbl 07498962 · doi:10.1080/10618600.2017.1302882
[28] Rosenblatt, M., Remarks on some nonparametric estimates of a density function, Ann. Math. Stat., 27, 3, 832-837 (1956) · Zbl 0073.14602 · doi:10.1214/aoms/1177728190
[29] Sahu, Sk; Dey, Dk; Branco, Md, A new class of multivariate skew distributions with applications to Bayesian regression models, Can. J. Stat., 31, 2, 129-150 (2008) · Zbl 1039.62047 · doi:10.2307/3316064
[30] Shestopaloff, AY., Neal, RM.: On Bayesian inference for the M/G/1 queue with efficient MCMC sampling. ArXiv preprint arXiv:1401.5548 (2014)
[31] Silverman, Bw, Density Estimation for Statistics and Data Analysis (2018), Abingdon: Routledge, Abingdon
[32] Sisson, Sa; Fan, Y.; Beaumont, M., Handbook of Approximate Bayesian Computation (2018), Boca Raton: Chapman and Hall/CRC, Boca Raton
[33] Sklar, M., Fonctions de répartition à n dimensions et leurs marges, Inst. Stat. Univ. Paris, 8, 229-231 (1959) · Zbl 0100.14202
[34] Terrell, Gr; Scott, Dw, Variable kernel density estimation, Ann. Stat., 20, 3, 1236-1265 (1992) · Zbl 0763.62024 · doi:10.1214/aos/1176348768
[35] Warton, Di, Penalized normal likelihood and ridge regularization of correlation and covariance matrices, J. Am. Stat. Assoc., 103, 481, 340-349 (2008) · Zbl 1471.62362 · doi:10.1198/016214508000000021
[36] Wood, Sn, Statistical inference for noisy nonlinear ecological dynamic systems, Nature, 466, 1102-1107 (2010) · doi:10.1038/nature09319
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.