Markov chain Monte Carlo without likelihoods
Open accessThe DOI matches its Crossref record; 20 of the 20 references it cites were found in a registry or an index.
Abstract
Many stochastic simulation approaches for generating observations from a posterior distribution depend on knowing a likelihood function. However, for many complex probability models, such likelihoods are either impossible or computationally prohibitive to obtain. Here we present a Markov chain Monte Carlo method for generating observations from a posterior distribution without the use of likelihoods. It can also be used in frequentist applications, in particular for maximum-likelihood estimation. The approach is illustrated by an example of ancestral inference in population genetics. A number of open problems are highlighted in the discussion.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“The greatest computational advantage of the original ABC rejection algorithm over Monte Carlo based ABC methods is that it generates independent samples and therefore readily lends itself to ‘embarrassingly’ parallel computation [86]. The computation time thus scales linearly to the number of processors available.”
From Integrating Crop Growth Models with Whole Genome Prediction through Approximate Bayesian Computation (2015) · cites this paper as [86]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“If ϵ is sufficiently small then this distribution will be a good approximation for the ‘true’ posterior distribution, P(θ|D0).”
From Simulation-based model selection for dynamical systems in systems and population biology (2010) · cites this paper as Marjoram et al., 2003
Experimental — accuracy not yet measuredThe check wasn’t confident either way, so it gives no verdict — never disputed.
In the citing paper
“Moreover, related ideas are already applied in the context of generalized linear mixed models [16], unobserved data points [17].”
From Likelihood based observability analysis and confidence intervals for predictions of dynamic models (2012) · cites this paper as [17]
Experimental — accuracy not yet measuredIn the citing paper
“The price to pay is the generation of dependent outcomes, and the risk of getting stuck in regions of low probability of the state space for long periods of time.”
From Reverse engineering and identification in systems biology: strategies, perspectives and challenges (2014) · cites this paper as [193]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Equation of State Calculations by Fast Computing Machines
Metropolis, N., Rosenbluth, A. W., Rosenbluth, M. N., Teller, A. H., & Teller, E. (1953) · The Journal of Chemical Physics · doi:10.1063/1.1699114 · Cited by 37,192
Gelman, A., Carlin, J. B., Stern, H. S., & Rubin, D. B. (2003) · doi:10.1201/9780429258480 · Cited by 27,276
Monte Carlo sampling methods using Markov chains and their applications.
Hastings, W. K. (1970) · Biometrika · doi:10.1093/biomet/57.1.97 · Cited by 15,195
Markov Chain Monte Carlo in Practice
(1995) · Chapman & Hall/CRC eBooks · doi:10.1201/b14835 · Cited by 7,757
Approximate Bayesian computation in population genetics
Beaumont, M. A., Zhang, W., & Balding, D. J. (2002) · Genetics · doi:10.1093/genetics/162.4.2025 · Cited by 2,932
Ripley, B. D. (1987) · Wiley Series in Probability and Statistics · doi:10.1002/9780470316726 · Cited by 2,241
Bayes and Empirical Bayes Methods for Data Analysis
(2001) · Technometrics · doi:10.1198/tech.2001.s608 · Cited by 2,098
On the genealogy of large populations
Kingmán, J. F. C. (1982) · Journal of Applied Probability · doi:10.1017/s0021900200034446 · Cited by 1,628
Population growth of human Y chromosomes: a study of Y chromosome microsatellites
Pritchard, J. K., Seielstad, M. T., Perez-Lezaun, A., & Feldman, M. W. (1999) · Molecular Biology and Evolution · doi:10.1093/oxfordjournals.molbev.a026091 · Cited by 1,080
Bayes and Empirical Bayes Methods for Data Analysis, Second Edition
Carlin, B., & Louis, T. (2000) · Chapman & Hall/CRC Texts in Statistical Science · doi:10.1201/9781420057669 · Cited by 959
Research works printed but not linked: 4
- Kingman, J. F. C. (1982) J. Appl. Prob. 19A, 27–43. [Google Scholar] · doi:10.2307/3213548 · Paper page
- Stephens, M. (2001) in Handbook of Statistical Genetics, eds. Balding, D. J., Bishop, M. J. & Cannings, C. (Wiley, New York), pp. 213–238.
- Plagnol, V. & Tavaré, S. (2004) in Proceedings of the 5th International Conference on Monte Carlo and Quasi-Monte Carlo Methods in Scientific Computing, ed. Niederreiter, H. (Springer, Heidelberg).
- Cabrera, J. & Yohai, V. J. (1999) A New Computational Approach for Bayesian and Robust Bayesian Statistical Analysis, www.rci.rutgers.edu/~cabrera/pap/vic.pdf, preprint.
Cited by
Machine learning and the physical sciences
Carleo, G., Cirac, I., Cranmer, K., Daudet, L., Schuld, M., Tishby, N., Vogt-Maranto, L., & Zdeborová, L. (2019) · Reviews of Modern Physics · doi:10.1103/revmodphys.91.045002 · Cited by 2,417
Probabilistic machine learning and artificial intelligence
Ghahramani, Z. (2015) · Nature · doi:10.1038/nature14541 · Cited by 1,995
“The fact that defining such universal inference algorithms for computer programs is even possible is somewhat surprising, but it is related to the generality of certain key ideas from sampling such as rejection sampling, sequential Monte Carlo methods 25 and ‘approximate Bayesian computation’ 49 .”
BackgroundToni, T., Welch, D., Strelkowa, N., Ipsen, A., & Stumpf, M. P. H. (2009) · Journal of The Royal Society Interface · doi:10.1098/rsif.2008.0172 · Cited by 1,498
“In ABC methods, the evaluation of the likelihood is replaced by a simulation-based procedure (Pritchard et al. 1999; Beaumont et al. 2002; Marjoram et al. 2003; Sisson et al. 2007).”
BackgroundMethodInfluentialApproximate Bayesian Computation (ABC) in practice
Csilléry, K., Blum, M. G. B., Gaggiotti, O. E., & François, O. (2010) · Trends in Ecology & Evolution · doi:10.1016/j.tree.2010.04.001 · Cited by 1,302
Approximate Bayesian Computation in Evolution and Ecology
Beaumont, M. A. (2010) · Annual Review of Ecology, Evolution, and Systematics · doi:10.1146/annurev-ecolsys-102209-144621 · Cited by 1,114
“As noted by Becquet & Przeworski (2007), there is a close relationship between the algorithm by Marjoram et al. (2003) and another algorithm that has become known as the pseudo-marginal method (Andrieu & Roberts 2009) introduced by Beaumont (2003).”
BackgroundMethodInfluentialProgramming With Models: Writing Statistical Algorithms for General Model Structures With NIMBLE
de Valpine, P., Turek, D., Paciorek, C. J., Anderson-Bergman, C., Lang, D. T., & Bodik, R. (2017) · Journal of Computational and Graphical Statistics · doi:10.1080/10618600.2016.1172487 · Cited by 998
“…et al., 2014); (vi) “likelihood free” or “plug-and-play” methods such as synthetic likelihood (Wood, 2010), approximate Bayesian computation (ABC; Marjoram et al., 2003), or iterated filtering (Ionides et al., 2006); (vii) parametric bootstrapping of any of the above ideas; or (viii) the same…”
MethodThe frontier of simulation-based inference
Cranmer, K., Brehmer, J., & Louppe, G. (2020) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.1912789117 · Cited by 1,032
“(12), and highlight the combination with MCMC (13) and Sequential Monte Carlo (14).”
InfluentialHuang, W., Takebayashi, N., Qi, Y., & Hickerson, M. J. (2011) · BMC Bioinformatics · doi:10.1186/1471-2105-12-1 · Cited by 951
“MTML-msBayes implements hierarchical ABC by way of a standard rejection/acceptance algo-rithm [10,26-30] followed by an optional transformation step.”
MethodApproximate Bayesian computational methods
Marin, J.-M., Pudlo, P., Robert, C. P., & Ryder, R. J. (2012) · Statistics and Computing · doi:10.1007/s11222-011-9288-2 · Cited by 818
“The unavailable likelihood is estimated by the indicator of the tolerance zone or an average of indicators as in Marjoram et al. (2003).”
BackgroundSequential Monte Carlo without likelihoods
Sisson, S. A., Fan, Y., & Tanaka, M. M. (2007) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.0607208104 · Cited by 856
abc: an R package for approximate Bayesian computation (ABC)
Csilléry, K., François, O., & Blum, M. G. B. (2012) · Methods in Ecology and Evolution · doi:10.1111/j.2041-210x.2011.00179.x · Cited by 707
“Finally, we note that alternative algorithms exist that sample from an updated distribution that is closer in shape to the pos- terior than to the prior (Marjoram et al. 2003; Beaumont et al. 2009; Wegmann et al. 2010).”
BackgroundCornuet, J.-M., Santos, F., Beaumont, M. A., Robert, C. P., Marin, J.-M., Balding, D. J., Guillemaud, T., & Estoup, A. (2008) · Bioinformatics · doi:10.1093/bioinformatics/btn514 · Cited by 690
“Consequently, a line of research including the works of Tavaré et al. (1997), Weiss and von Haeseler (1998), Pritchard et al. (1999) and Marjoram et al. (2003) developed a new approach termed Approximate Bayesian Computation (or ABC) by Beaumont et al. (2002).”
MethodApproximate Bayesian Computation
Sunnåker, M., Busetto, A. G., Numminen, E., Corander, J., Foll, M., & Dessimoz, C. (2013) · PLoS Computational Biology · doi:10.1371/journal.pcbi.1002803 · Cited by 623
“Methods for the identification of summary statistics that could also simultaneously assess the influence on the approximation of the posterior would be of substantial value [33].”
BackgroundMethodAdaptive approximate Bayesian computation
Beaumont, M. A., Cornuet, J.-M., Marin, J.-M., & Robert, C. P. (2009) · Biometrika · doi:10.1093/biomet/asp052 · Cited by 614
“While Sisson et al.'s (2007) algorithm induces biased weights, with a visible impact on the quality of the approximation, we have shown that the same Markov transition kernels and thus the same computing power can be used to produce an unbiased scheme.”
BackgroundMethodInfluentialStatistical evaluation of alternative models of human evolution
Fagundes, N. J. R., Ray, N., Beaumont, M., Neuenschwander, S., Salzano, F. M., Bonatto, S. L., & Excoffier, L. (2007) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.0708280104 · Cited by 589
Fearnhead, P., & Prangle, D. (2012) · Journal of the Royal Statistical Society Series B: Statistical Methodology · doi:10.1111/j.1467-9868.2011.01010.x · Cited by 567
“An alternative Monte Carlo procedure for implementing ABC is based on Markov chain Monte Carlo (MCMC) sampling (Marjoram et al., 2003; Bortot et al., 2007) and given in algorithm 2 (Table 2).”
MethodThe Bayesian revolution in genetics
Beaumont, M. A., & Rannala, B. (2004) · Nature Reviews Genetics · doi:10.1038/nrg1318 · Cited by 525
“An example of the latter is what has come to be known as ' APPROXIMATE BAYESIAN COMPUTATION ' (ABC”
BackgroundAn adaptive sequential Monte Carlo method for approximate Bayesian computation
Del Moral, P., Doucet, A., & Jasra, A. (2012) · Statistics and Computing · doi:10.1007/s11222-011-9271-y · Cited by 505
“(2009), Sisson et al. (2009), Toni et al. (2009). However, it is worth mentioning that all the previous SMC algorithms proposed for ABC and the one presented here require simulating a number of pseudo-observations proportional to N .”
BackgroundMethodInfluentialCornuet, J.-M., Ravigné, V., & Estoup, A. (2010) · BMC Bioinformatics · doi:10.1186/1471-2105-11-401 · Cited by 479
“Whereas several studies have now shown that parameter posterior distributions inferred by ABC are similar to those provided by full-likelihood Bayesian approaches [16-19], the approach is still in its infancy and continues to evolve, and to be improved (reviewed in [4,5,20,21]).”
BackgroundABC as a flexible framework to estimate demography over space and time: some cons, many pros
BERTORELLE, G., BENAZZO, A., & MONA, S. (2010) · Molecular Ecology · doi:10.1111/j.1365-294x.2010.04690.x · Cited by 464
“used, although the importance of this step was already recognized since the formal introduction of ABC (Beaumont et al. 2002; Marjoram et al. 2003).”
BackgroundMethodInfluentialNon-linear regression models for Approximate Bayesian Computation
Blum, M. G. B., & François, O. (2010) · Statistics and Computing · doi:10.1007/s11222-009-9116-0 · Cited by 453
“The second class of algorithms mimics Markov chain Monte Carlo methods (MCMC, Robert and Casella 2004), embedding simulations from the implicit model in the updating step of the stochastic algorithm (Marjoram et al. 2003; Bortot et al. 2007).”
MethodAbbasi, R., Ackermann, M., Adams, J., Aguilar, J. A., Ahlers, M., Ahrens, M., Alispach, C., Alves, A. A., Amin, N. M., Andeen, K., et al. (2021) · Physical Review D · doi:10.1103/physrevd.104.022002 · Cited by 363
“After the ABC step, we marginalize over other parameters to obtain the event direction’s posterior distribution.”
MethodInfluentialLearning in Implicit Generative Models
Mohamed, S., & Lakshminarayanan, B. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1610.03483 · Cited by 157
“The two other approaches are: moment matching and ABC-MCMC (Marjoram et al., 2003), which has been successful for many problems with moderate dimension; and the nat-ural choice of gradient-free optimisation methods, which include familiar tools such as Bayesian optimisation (Gut-mann and Corander, 2016), evolutionary…”
MethodComputer programs for population genetics data analysis: a survival guide
Excoffier, L., & Heckel, G. (2006) · Nature Reviews Genetics · doi:10.1038/nrg1904 · Cited by 408
“However, some approximate Bayesian methods, which rely on summary statistics instead of likelihoods, have been developed recently to deal with more complex model”
MethodABCtoolbox: a versatile toolkit for approximate Bayesian computations
Wegmann, D., Leuenberger, C., Neuenschwander, S., & Excoffier, L. (2010) · BMC Bioinformatics · doi:10.1186/1471-2105-11-116 · Cited by 404
“Additional improvements have been proposed to more efficiently explore the parameter space, including several variations of likelihood-free MCMC [6-9] or Particle-based samplers [10-12].”
MethodInfluentialStatistical inference for stochastic simulation models - theory and application
Hartig, F., Calabrese, J. M., Reineking, B., Wiegand, T., & Huth, A. (2011) · Ecology Letters · doi:10.1111/j.1461-0248.2011.01640.x · Cited by 384
“The ABC approach is asymptotically exact, meaning that it will, for suitable distance measures and sufficient summary statistics, reproduce the true shape of the likelihood function in the limit of e fi 0 and N fi ¥, N being the sampling effort (Marjoram et al. 2003).”
BackgroundMethodInfluential3-D Structural geological models: Concepts, methods, and uncertainties
Wellmann, F., & Caumon, G. (2018) · Advances in Geophysics · doi:10.1016/bs.agph.2018.09.001 · Cited by 347
“Approximate Bayesian Computation in Popu- lation Genetics.”
MethodNachman, M. W., & Payseur, B. A. (2012) · Philosophical Transactions of the Royal Society B: Biological Sciences · doi:10.1098/rstb.2011.0249 · Cited by 376
A Comparative Review of Dimension Reduction Methods in Approximate Bayesian Computation
Blum, M. G. B., Nunes, M. A., Prangle, D., & Sisson, S. A. (2013) · Statistical Science · doi:10.1214/12-sts406 · Cited by 353
“A simple rejection-sampling algorithm to achieve this was proposed by Pritchard et al. (1999) (see also Marjoram et al. 2003), which produces draws from p(θ, s|sobs).”
BackgroundMethodInfluentialBeven, K., & Binley, A. (2014) · Hydrological Processes · doi:10.1002/hyp.10082 · Cited by 336
Efficient Approximate Bayesian Computation Coupled With Markov Chain Monte Carlo Without Likelihood
Wegmann, D., Leuenberger, C., & Excoffier, L. (2009) · Genetics · doi:10.1534/genetics.109.102509 · Cited by 348
Cancer Evolution: Mathematical Models and Computational Inference
Beerenwinkel, N., Schwarz, R. F., Gerstung, M., & Markowetz, F. (2015) · Systematic Biology · doi:10.1093/sysbio/syu081 · Cited by 345
“Statistical inference in the coalescent can be computationally demanding and several approximations to ML or Bayesian estimation have been proposed, most notably Approximate Bayesian Computation, which avoids evaluation of the likelihood function (Marjoram et al. 2003; Haccou et al. 2005).”
MethodA direct characterization of human mutation based on microsatellites
Sun, J. X., Helgason, A., Masson, G., Ebenesersdóttir, S. S., Li, H., Mallick, S., Gnerre, S., Patterson, N., Kong, A., Reich, D., et al. (2012) · Nature Genetics · doi:10.1038/ng.2398 · Cited by 340
Fast $ε$-free Inference of Simulation Models with Bayesian Conditional Density Estimation
Papamakarios, G., & Murray, I. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1605.06376 · Cited by 133
“We compare to three ABC baselines: (a) rejection ABC [25], where parameters are proposed from the prior and are accepted if (cid:107) x − x o (cid:107) < (cid:15) ; (b) MCMC-ABC [16] with a spherical Gaussian proposal, whose variance we manually tuned separately in each case for best performance; and (c) SMC-ABC [3],…”
MethodResultInfluentialLack of confidence in approximate Bayesian computation model choice
Robert, C. P., Cornuet, J.-M., Marin, J.-M., & Pillai, N. S. (2011) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.1102900108 · Cited by 317
A tutorial on approximate Bayesian computation
Turner, B. M., & Van Zandt, T. (2012) · Journal of Mathematical Psychology · doi:10.1016/j.jmp.2012.02.005 · Cited by 293
Detecting Selection in Population Trees: The Lewontin and Krakauer Test Extended
Bonhomme, M., Chevalet, C., Servin, B., Boitard, S., Abdallah, J., Blott, S., & SanCristobal, M. (2010) · Genetics · doi:10.1534/genetics.104.117275 · Cited by 285
“…when the population number is not too large, alternative methods such as approximate Bayesian computation (ABC) methods (Beaumont et al. 2002; Marjoram et al. 2003) could be considered, as they potentially deal with more summary statistics than only Detecting Selection in Population Trees…”
MethodTraining deep neural density estimators to identify mechanistic models of neural dynamics
Gonçalves, P. J., Lueckmann, J.-M., Deistler, M., Nonnenmacher, M., Öcal, K., Bassetto, G., Chintaluri, C., Podlaski, W. F., Haddad, S. A., Vogels, T. P., et al. (2020) · eLife · doi:10.7554/elife.56261 · Cited by 239
Golightly, A., & Wilkinson, D. J. (2011) · Interface Focus · doi:10.1098/rsfs.2011.0047 · Cited by 282
Statistical inferences in phylogeography
NIELSEN, R., & BEAUMONT, M. A. (2009) · Molecular Ecology · doi:10.1111/j.1365-294x.2008.04059.x · Cited by 272
“Typically, the likelihood surface varies rapidly with increasing distance from the target (Beaumont et al. 2002), and a variety of techniques have been developed to try to correct for this (Beaumont et al. 2002; Marjoram et al. 2003; Sisson et al. 2007).”
BackgroundSequential Neural Likelihood: Fast Likelihood-free Inference with Autoregressive Flows
Papamakarios, G., Sterratt, D. C., & Murray, I. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1805.07226 · Cited by 67
“A basic implementation of ABC would simulate parameters from the prior; more sophisticated variants such as Markov Chain Monte Carlo ABC [47, 51, 69] and Sequential Monte Carlo ABC [6, 10, 70, 76] guide future simulations based on previously accepted parameters.”
MethodBayesian Analysis of an Admixture Model With Mutations and Arbitrarily Linked Markers
Excoffier, L., Estoup, A., & Cornuet, J.-M. (2005) · Genetics · doi:10.1534/genetics.104.036236 · Cited by 264
“This approach does not require the computation of likelihoods, but simply relies on the comparison of summary statistics computed on observed data with those computed on data simulated under a model for which the parameters of interest are known (Beaumont et al. 2002; Marjoram et al. 2003).”
MethodIntegrating Crop Growth Models with Whole Genome Prediction through Approximate Bayesian Computation
Technow, F., Messina, C. D., Totir, L. R., & Cooper, M. (2015) · PLOS ONE · doi:10.1371/journal.pone.0130855 · Cited by 265
“Another interesting development isMCMC-ABC, which incorporates ABC with the Metropolis-Hastings algorithm [86].”
MethodSalmonella persisters promote the spread of antibiotic resistance plasmids in the gut
Bakkeren, E., Huisman, J. S., Fattinger, S. A., Hausmann, A., Furter, M., Egli, A., Slack, E., Sellin, M. E., Bonhoeffer, S., Regoes, R. R., et al. (2019) · Nature · doi:10.1038/s41586-019-1521-8 · Cited by 255
Haddrill, P. R., Thornton, K. R., Charlesworth, B., & Andolfatto, P. (2005) · Genome Research · doi:10.1101/gr.3541005 · Cited by 257
“…on summary statistics of the data (sample size, alignment length, diversity, , number of segregating sites, S, the number of haplotypes, K, and the minimum number of recombination events in the sample, Rm) and rejection sampling (cf., Pritchard et al. 1999 and Algorithm D of Marjoram et al. 2003).”
BackgroundMethodSimulation-based model selection for dynamical systems in systems and population biology
Toni, T., & Stumpf, M. P. H. (2010) · Bioinformatics · doi:10.1093/bioinformatics/btp619 · Cited by 260
“Contact: ttoni@imperial.ac.uk; m.stumpf@imperial.ac.uk Supplementary information: Supplementary data are available at Bioinformatics online.”
BackgroundDeep Learning for Population Genetic Inference
Sheehan, S., & Song, Y. S. (2016) · PLOS Computational Biology · doi:10.1371/journal.pcbi.1004845 · Cited by 245
“In this vein, it would be interesting to use an approach such as ABC MCMC [24] to simulate data, and Table 9.”
BackgroundMethodThornton, K., & Andolfatto, P. (2006) · Genetics · doi:10.1534/genetics.105.048223 · Cited by 256
“Approximate Bayesian inference: We took an approximate Bayesian approach (Pritchard et al. 1999; Beaumont et al. 2002; Marjoram et al. 2003; Przeworski 2003) to obtain k independent samples from the joint posterior distribution of model parameters by rejection sampling.”
MethodSimulating Trees with a Fixed Number of Extant Species
Stadler, T. (2011) · Systematic Biology · doi:10.1093/sysbio/syr029 · Cited by 264
“The simulation tools developed in this paper provide the first step toward implementing an approximate Bayesian computation (ABC) approach (Marjoram et al. 2003) for estimating rates under the EBDP model.”
MethodSemiparametric regression during 2003–2007
Ruppert, D., Wand, M. P., & Carroll, R. J. (2009) · Electronic Journal of Statistics · doi:10.1214/09-ejs525 · Cited by 250
“Other recent developments in Monte Carlo methods that lend themselves to semiparametric regression applications include slice sampling with ‘stepping out’ (Neal, 2003) and approximate Bayesian computation (e.g. Beaumont et al. 2002; Marjoram et al. 2003; Sisson et al. 2007).”
MethodApproximate Bayesian computation (ABC) gives exact results under the assumption of model error
Wilkinson, R. D. (2013) · Statistical Applications in Genetics and Molecular Biology · doi:10.1515/sagmb-2013-0010 · Cited by 247
A survey of Monte Carlo methods for parameter estimation
Luengo, D., Martino, L., Bugallo, M., Elvira, V., & Särkkä, S. (2020) · EURASIP Journal on Advances in Signal Processing · doi:10.1186/s13634-020-00675-6 · Cited by 199
Approximate Bayesian Computation: A Nonparametric Perspective
Blum, M. G. B. (2010) · Journal of the American Statistical Association · doi:10.1198/jasa.2010.tm09448 · Cited by 239
“Such adaptive ABC algorithms encompass ABC-MCMC algorithms (Marjoram et al. 2003, Bortot et al. 2007) and ABC sequential Monte Carlo samplers (Sisson et al. 2007, Beaumont et al. 2009, Toni et al. 2009).”
Drovandi, C. C., & Pettitt, A. N. (2011) · Biometrics · doi:10.1111/j.1541-0420.2010.01410.x · Cited by 230
“Marjoram et al. (2003) attempted to overcome this problem by developing an MCMC algorithm without likelihoods.”
BackgroundMethodInfluentialReverse engineering and identification in systems biology: strategies, perspectives and challenges
Villaverde, A. F., & Banga, J. R. (2014) · Journal of The Royal Society Interface · doi:10.1098/rsif.2013.0505 · Cited by 229
“[193] presented a MCMC algorithm (ABC MCMC) that accepts observations more frequently and does not require the computation of likelihoods.”
MethodFundamentals and Recent Developments in Approximate Bayesian Computation
Lintusaari, J., Gutmann, M. U., Dutta, R., Kaski, S., & Corander, J. (2016) · Systematic Biology · doi:10.1093/sysbio/syw077 · Cited by 219
“For completeness, we state below the algorithms for MCMC ABC and SMC ABC by Marjoram et al. (2003) and Beaumont et al. (2009), respectively.”
MethodInfluentialMining gold from implicit models to improve likelihood-free inference
Brehmer, J., Louppe, G., Pavez, J., & Cranmer, K. (2020) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.1915980117 · Cited by 159
Bayesian inference for exponential random graph models
Caimo, A., & Friel, N. (2011) · Social Networks · doi:10.1016/j.socnet.2010.09.004 · Cited by 224
“Note that this algo-rithm has some similarities with another likelihood-free method called Approximate Bayesian Computation (ABC) (see Beaumont et al. (2002) and Marjoram et al. (2003)).”
MethodBayesian Optimization for Likelihood-Free Inference of Simulator-Based Statistical Models
Gutmann, M. U., & Corander, J. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1501.03291 · Cited by 208
“, 1993), “approximate Bayesian computation” (ABC) in genetics (Beaumont et al., 2002; Marjoram et al., 2003; Sisson et al., 2007), or the “synthetic likelihood” approach in ecology (Wood, 2010).”
BackgroundDating primate divergences through an integrated analysis of palaeontological and molecular data
Wilkinson, R. D., Steiper, M. E., Soligo, C., Martin, R. D., Yang, Z., & Tavaré, S. (2011) · Systematic Biology · doi:10.1093/sysbio/syq054 · Cited by 218
“Rejection-based ABC cannot handle large numbers of parameters, because repeatedly sampling from the prior distribution becomes too inefficient in high dimensions.”
MethodInfluentialSelf-citationLiepe, J., Kirk, P., Filippi, S., Toni, T., Barnes, C. P., & Stumpf, M. P. H. (2014) · Nature Protocols · doi:10.1038/nprot.2014.025 · Cited by 217
“It is usually impossible to write down an expression for the poste rior distribution analytically; in these cases, it is necessary to use computational approaches, such as Markov chain Monte Carlo (MCMC) techniques 15 .”
MethodInfluentialUsing Genomic Data to Infer Historic Population Dynamics of Nonmodel Organisms
Beichman, A. C., Huerta-Sanchez, E., & Lohmueller, K. E. (2018) · Annual Review of Ecology, Evolution, and Systematics · doi:10.1146/annurev-ecolsys-110617-062431 · Cited by 211
Nestedness in networks: A theoretical model and some applications
König, M. D., Tessone, C. J., & Zenou, Y. (2014) · Theoretical Economics · doi:10.3982/te1348 · Cited by 222
“We then estimate the main parameters θ ≡ (α ζ) of our model by using the likelihoodfree Markov chain Monte Carlo (LF-MCMC) algorithm suggested by Marjoram et al. (2003). The purpose of this algorithm is to estimate the parameter vector θ of our model on the basis of the summary statistics S ≡ (S1 S2 S3)n×3, where S1 ≡…”
InfluentialApproximate Bayesian Computation
Beaumont, M. A. (2019) · Annual Review of Statistics and Its Application · doi:10.1146/annurev-statistics-030718-105212 · Cited by 183
Robust Bayesian Inference via Coarsening
Miller, J. W., & Dunson, D. B. (2019) · Journal of the American Statistical Association · doi:10.1080/01621459.2018.1469995 · Cited by 159
“The c-posterior is mathematically equivalent to the type of posterior approximation resulting from approximate Bayesian computation (ABC) (Tavaré et al., 1997; Marjoram et al., 2003; Beaumont et al., 2002; Wilkinson, 2013)—indeed, ABC provided part of our inspiration for considering this form of posterior.”
BackgroundMethodModern computational approaches for analysing molecular genetic variation data
Marjoram, P., & Tavaré, S. (2006) · Nature Reviews Genetics · doi:10.1038/nrg1961 · Cited by 198
“In the absence of a sufficient statistic, we rely on intuition to choose S , and then, perhaps, calibrate answers for a simpler form of the model from which we can find the exact posterior distributio”
MethodSelf-citationIn defence of model-based inference in phylogeography
BEAUMONT, M. A., NIELSEN, R., ROBERT, C., HEY, J., GAGGIOTTI, O., KNOWLES, L., ESTOUP, A., PANCHAL, M., CORANDER, J., HICKERSON, M., et al. (2010) · Molecular Ecology · doi:10.1111/j.1365-294x.2009.04515.x · Cited by 185
“For instance, Marjoram et al. (2003) developed a Markov chain Monte Carlo (MCMC) ABC approach, improving the sampling efficiency of conventional ABC, which must otherwise explore sometimes very wide priors while posterior distributions may only occupy a narrow region of parameter space.”
Distinguishing between selective sweeps from standing variation and from a de novo mutation
Peter, B. M., Huerta-Sanchez, E., & Nielsen, R. (2012) · PLoS Genetics · doi:10.1371/journal.pgen.1003011 · Cited by 193
Impact and cost-effectiveness of new tuberculosis vaccines in low- and middle-income countries
Knight, G. M., Griffiths, U. K., Sumner, T., Laurence, Y. V., Gheorghe, A., Vassall, A., Glaziou, P., & White, R. G. (2014) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.1404386111 · Cited by 190
Inference in Epidemic Models without Likelihoods
McKinley, T., Cook, A. R., & Deardon, R. (2009) · The International Journal of Biostatistics · doi:10.2202/1557-4679.1171 · Cited by 183
Flexible statistical inference for mechanistic models of neural dynamics
Lueckmann, J.-M., Gonçalves, P. J., Bassetto, G., Öcal, K., Nonnenmacher, M., & Macke, J. H. (2017) · Lirias · doi:10.48550/arxiv.1711.01861 · Cited by 129
“More efficient variants make use of a Markov-Chain Monte-Carlo [23, 24] or Sequential Monte-Carlo (SMC) samplers [25, 26].”
MethodShafii, M., & Tolson, B. A. (2015) · Water Resources Research · doi:10.1002/2014wr016520 · Cited by 163
“…[2013] was the first to apply the approximate Bayesian computation (ABC) concept from the statistics literature [e.g., Del Moral et al., 2012; Marjoram et al., 2003; Sisson et al., 2007] to hydrology as a tool for using hydrological signatures in diagnostic model evaluation, aiming to…”
MethodToward diagnostic model calibration and evaluation: Approximate Bayesian computation
Vrugt, J. A., & Sadegh, M. (2013) · Water Resources Research · doi:10.1002/wrcr.20354 · Cited by 165
“The premise behind ABC is that should be a sample from the posterior distribution as long as the distance between the observed and simulated data, hereafter referred to as ~Y;Y ð Þ is less than some small positive value, [Marjoram et al., 2003; Sisson et al., 2007].”
BackgroundMethodBayesian Computation and Model Selection Without Likelihoods
Leuenberger, C., & Wegmann, D. (2010) · Genetics · doi:10.1534/genetics.109.109058 · Cited by 175
“Marjoram et al. (2003) develop a variant of the classical Metropolis–Hastings algorithm (termed ABC–MCMC in Sisson et al. 2007), which allows them to sample directly from the truncated prior peðuÞ.”
BackgroundMethodInfluentialLimitations on low rank approximations for covariance matrices of spatial data
Stein, M. L. (2014) · Spatial Statistics · doi:10.1016/j.spasta.2013.06.003 · Cited by 180
Maximizing the information content of experiments in systems biology
Liepe, J., Filippi, S., Komorowski, M., & Stumpf, M. P. H. (2013) · PLoS Computational Biology · doi:10.1371/journal.pcbi.1002888 · Cited by 170
“Finding the posterior, p(hDx), is usually achieved by means of powerful (if costly) computational algorithms such as Markov chain Monte Carlo (MCMC) and sequential Monte Carlo (SMC) methods, which also exist in the approximate Bayesian computation (ABC) framework [15,35,36].”
BayesSim: Adaptive Domain Randomization Via Probabilistic Inference for Robotics Simulators
Ramos, F., Possas, R., & Fox, D. (2019) · Robotics: Science and Systems XV · doi:10.15607/rss.2019.xv.029 · Cited by 111
Gaia Early Data Release 3: Structure and properties of the Magellanic Clouds
Gaia Collaboration, Luri, X., Chemin, L., Clementini, G., Delgado, H. E., McMillan, P. J., Romero-Gómez, M., Balbinot, E., Castro-Ginard, A., Mor, R., et al. (2021) · Astronomy & Astrophysics · doi:10.1051/0004-6361/202039588 · Cited by 143
Bayesian Solution Uncertainty Quantification for Differential Equations
Chkrebtii, O., Campbell, D. A., Calderhead, B., & Girolami, M. (2016) · Bayesian Analysis · doi:10.1214/16-ba1017 · Cited by 191
“Such partially likelihood-free Markov chain Monte Carlo methods (Marjoram et al., 2003) are widely used for inference on stochastic differential equations (e.g., Golightly and Wilkinson, 2011).”
Microevolution of Helicobacter pylori during Prolonged Infection of Single Hosts and within Families
Morelli, G., Didelot, X., Kušećek, B., Schwarz, S., Bahlawane, C., Falush, D., Suerbaum, S., & Achtman, M. (2010) · PLoS Genetics · doi:10.1371/journal.pgen.1001036 · Cited by 171
“[56].”
Adaptive approximate Bayesian computation for complex models
Lenormand, M., Jabot, F., & Deffuant, G. (2013) · Computational Statistics · doi:10.1007/s00180-013-0428-3 · Cited by 155
“Beaumont et al. (2009); Toni et al. (2009) proposed a new algorithm, called Population Monte Carlo ABC in Beaumont et al. (2009) and hereafter called PMC.”
MethodInfluentialOn the Statistical Calibration of Physical Models
Sargsyan, K., Najm, H. N., & Ghanem, R. (2015) · International Journal of Chemical Kinetics · doi:10.1002/kin.20906 · Cited by 152
A generalized, likelihood-free method for posterior estimation
Turner, B. M., & Sederberg, P. B. (2013) · Psychonomic Bulletin & Review · doi:10.3758/s13423-013-0530-0 · Cited by 176
On uncertainty quantification in hydrogeology and hydrogeophysics
Linde, N., Ginsbourger, D., Irving, J., Nobile, F., & Arnaud Doucet, B. (2017) · Advances in Water Resources · doi:10.1016/j.advwatres.2017.10.014 · Cited by 156
Tanaka, M. M., Francis, A. R., Luciani, F., & Sisson, S. A. (2006) · Genetics · doi:10.1534/genetics.106.055574 · Cited by 152
“In these settings there are complex dependencies among individuals that can be simulated using the coalescent and related models, but likelihood functions are more difficult to write down (Marjoram et al. 2003).”
BackgroundMethodSisson, S. A., & Fan, Y. (2011) · Chapman & Hall/CRC eBooks · doi:10.1201/b10905-13 · Cited by 165
Effects of Linked Selective Sweeps on Demographic Inference and Model Selection
Schrider, D. R., Shanku, A. G., & Kern, A. D. (2016) · Genetics · doi:10.1534/genetics.116.190223 · Cited by 146
“…of a demographic model’s parameters through the use of a collection of population genetic summary statistics without specification of an explicit likelihood function (Tavaré et al. 1997; Pritchard et al. 1999; Beaumont et al. 2002; Marjoram et al. 2003; Excoffier et al. 2005; Wegmann et al. 2010).”
BackgroundThe state of plant population modelling in light of environmental change
Jeltsch, F., Moloney, K. A., Schurr, F. M., Köchy, M., & Schwager, M. (2008) · Perspectives in Plant Ecology Evolution and Systematics · doi:10.1016/j.ppees.2007.11.004 · Cited by 146
“Moreover, approximate Bayesian computation (Marjoram et al., 2003) helps in situations where the true likelihood cannot be evaluated.”
BackgroundJournal De La Société Française De Statistique
Caussinus, H., & de Falguerolles, A. (2005) · Wiley eBooks · doi:10.1002/0471667196.ess7017 · Cited by 141
Model criticism based on likelihood-free inference, with an application to protein network evolution
Ratmann, O., Andrieu, C., Wiuf, C., & Richardson, S. (2009) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.0807882106 · Cited by 139
“With respect to ( i ) , we compared ABC μ with ABC-MCMC (26), zoomABC-MCMC (12), and AUX-ABC (27) on the H. pylori PIN dataset ( SI Appendix , S1.”
MethodJabot, F., & Chave, J. (2008) · Ecology Letters · doi:10.1111/j.1461-0248.2008.01280.x · Cited by 138
Tsutaya, T., & Yoneda, M. (2013) · PLoS ONE · doi:10.1371/journal.pone.0072327 · Cited by 160
“Several algorithms have been proposed to increase the sampling efficiency, by introducing weighting with regression analysis [55,56], Markov chain Monte Carlo sampling [57], and sequential Monte Carlo (SMC) sampling [58–60].”
MethodSocial inheritance can explain the structure of animal social networks
Ilany, A., & Akçay, E. (2016) · Nature Communications · doi:10.1038/ncomms12084 · Cited by 144
“Methods such as approximate bayesian computation (Marjoram et al., 2003), coupled with dimensionality reduction techniques (Wegmann et al.”
MethodWitzel, G., Martinez, G., Hora, J., Willner, S. P., Morris, M. R., Gammie, C., Becklin, E. E., Ashby, M. L. N., Baganoff, F., Carey, S., et al. (2018) · The Astrophysical Journal · doi:10.3847/1538-4357/aace62 · Cited by 115
“The PMC-ABC sampler Approximate Bayesian Computation (ABC) is a useful computational algorithm for Bayesian parameter space exploration where explicit likelihood evaluations are either impossible or not feasible (Cameron & Pettitt 2012; Marjoram et al. 2003; Sisson et al. 2007).”
MethodInference for Stereological Extremes
Bortot, P., Coles, S., & Sisson, S. A. (2007) · Journal of the American Statistical Association · doi:10.1198/016214506000000988 · Cited by 135
“To address this difficulty, Marjoram et al. (2003) proposed two modifications to the basic algorithm.”
MethodLikelihood based observability analysis and confidence intervals for predictions of dynamic models
Kreutz, C., Raue, A., & Timmer, J. (2012) · BMC Systems Biology · doi:10.1186/1752-0509-6-120 · Cited by 134
“Discussion Existing approaches for prediction confidence intervals likeMCMC [26] or bootstrap procedures are based on forward evaluations of the model for many parameter values.”
MethodApproximate Bayesian Computation for infectious disease modelling
Minter, A., & Retkutė, R. (2019) · Epidemics · doi:10.1016/j.epidem.2019.100368 · Cited by 146
Robot Learning From Randomized Simulations: A Review
Muratore, F., Ramos, F., Turk, G., Yu, W., Gienger, M., & Peters, J. (2022) · Frontiers in Robotics and AI · doi:10.3389/frobt.2022.799893 · Cited by 128
“There exist plenty of variants for approximate Bayesian computation (Marjoram et al., 2003; Beaumont et al., 2009; Sunnåker et al., 2013) as well as studies on the design of low-dimensional summary statistics (Fearnhead and Prangle, 2012).”
BackgroundLikelihood-free inference via classification
Gutmann, M. U., Dutta, R., Kaski, S., & Corander, J. (2018) · Statistics and Computing · doi:10.1007/s11222-017-9738-6 · Cited by 90
“Approximate Bayesian computation (ABC) stems from research in genetics (Beaumont et al. 2002; Marjoram et al. 2003; Pritchard et al. 1999; Tavaré et al. 1997),while themethodof simulatedmoments (McFadden 1989; Pakes and Pollard 1989) and indirect inference (Gouriéroux et al.”
MethodOn optimality of kernels for approximate Bayesian computation using sequential Monte Carlo
Filippi, S., Barnes, C. P., Cornebise, J., & Stumpf, M. P. H. (2013) · Statistical Applications in Genetics and Molecular Biology · doi:10.1515/sagmb-2012-0069 · Cited by 154
“…three flavours: regression-adjusted ABC (Tallmon, 2004; Fagundes et al. , 2007; Blum & Fran¸cois, 2009), Markov chain Monte Carlo ABC schemes (Marjoram & Molitor, 2003; Ratmann et al. , 2007), and ABC implementing some variant of sequential importance sampling (SIS) or sequential Monte Carlo…”
BackgroundMethod
Metadata
- Authors
- Marjoram, P., Molitor, J., Plagnol, V., & Tavaré, S.
- Affiliations
- University of Southern California (US)
- Venue
- Proceedings of the National Academy of Sciences
- Publisher
- National Academy of Sciences
- Published
- 2003-12-08
- Volume
- 100
- Issue
- 26
- Pages
- 15324-15328
- ISSN
- 0027-8424 (ISSN-L)
- 1091-6490
- PMC
- PMC307566
- OpenAlex
- W2045973738
- Cited by
- 1,199
- References
- 20
- Citations by year
- 2026: 26 · 2025: 38 · 2024: 54 · 2023: 54 · 2022: 57 · 2021: 72 · 2020: 83 · 2019: 79 · 2018: 88 · 2017: 68 · 2016: 76 · 2015: 79 · 2014: 86 · 2013: 64 · 2012: 60
- FWCI
- 12.01
- Type
- journal article
- Language
- en
- Open access
- green
- www.ncbi.nlm.nih.gov
- Locations
- Proceedings of the National Academy of Sciences (publishedVersion)
- PubMed Central (publishedVersion)
- PubMed (publishedVersion)
- UCL Discovery (University College London) (submittedVersion)
- Funders
- National Institutes of Health (5r01gm058897-03, GM58897)
- National Institute of General Medical Sciences (GM 58897)
- Division of Mathematical Sciences
- Topics
- Physical Sciences › Mathematics › Statistics and Probability › Markov Chains and Monte Carlo Methods
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods and Bayesian Inference
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Genetics › Genetic diversity and population structure
- Keywords
- Markov chain Monte Carlo, simulation-based inference, posterior sampling, maximum likelihood estimation, population genetics
- MeSH terms
- Algorithms, Computer Simulation, DNA, DNA, Mitochondrial, Biological Evolution, Genetics, Population, Humans, Markov Chains, Models, Biological, Monte Carlo Method, Stochastic Processes, Likelihood Functions
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.