Abstract
Several structure-learning algorithms for staged trees, asymmetric extensions of Bayesian networks, have been proposed. However, these do not scale efficiently as the number of variables considered increases, a priori restricting the set of models, or they do not find comparable models to existing methods. Here, we define an alternative algorithm based on a totally-ordered hyperstage. We demonstrate how it can be used to obtain a quadratically-scaling structural learning algorithm for staged trees that restricts the model space a posteriori. Through comparative analysis, we show that through the ordering provided by the mean posterior distributions, we can outperform existing methods in computational time whilst providing comparable model scores. This method also enables us to learn more complex relationships than existing model selection techniques by expanding the model space. We illustrate how this can embellish inferences in a real study.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Barclay, L.M., Hutton, J.L., Smith, J.Q.: Refining a bayesian network using a chain event graph. Int. J. Approximate Reasoning 54(9), 1300–1309 (2013)
Barclay, L.M., Hutton, J.L., Smith, J.Q.: Chain event graphs for informed missingness. Bayesian Anal. 9(1) (2014)
Bunnin, F.O., Smith, J.Q.: A bayesian hierarchical model for criminal investigations. Bayesian Anal. 16(1), 1–30 (2021)
Carli, F., Leonelli, M., Riccomagno, E., Varando, G.: The r package stagedtrees for structural learning of stratified staged trees. J. Stat. Soft. 102(6), 1-30 (2022). https://doi.org/10.18637/jss.v102.i06
Collazo, R.A., Görgen, C., Smith, J.Q.: Chain event graphs. CRC Press (2018)
Collazo, R.A., Smith, J.Q.: A New Family of Non-Local Priors for Chain Event Graph Model Selection. Bayesian Analysis 11(4), 1165–1201 (2016). https://doi.org/10.1214/15-BA981. URLhttps://doi.org/10.1214/15-BA981
Cowell, R.G., Smith, J.Q.: Causal discovery through MAP selection of stratified chain event graphs. Electronic Journal of Statistics 8(1), 965–997 (2014)
Duarte, E., Solus, L.: Representation of Context-Specific Causal Mdels with Observational and Interventional data (2021). ArXiv prepint. arXiv:2101.09271
Görgen, C., Maraj, A., Nicklasson, L.: Staged tree models with toric structure. J. Sym. Comput. 113, 242–268 (2022)
Görgen, C., Smith, J.Q.: Equivalence classes of staged trees. Bernoulli 24(4A), 2676–2692 (2018). https://doi.org/10.3150/17-BEJ940
Hughes, C., Strong, P., Shenvi, A.: On bayesian dirichlet scores for staged trees and chain event graphs (2022). https://doi.org/10.48550/ARXIV.2206.15322, arXiv:2206.15322
Leonelli, M., Varando, G.: Highly efficient structural learning of sparse staged trees. In: Salmerón, A., Rumı, R. (eds.) Proceedings of The 11th International Conference on Probabilistic Graphical Models, Proceedings of Machine Learning Research, vol. 186, pp. 193–204. PMLR (2022). https://proceedings.mlr.press/v186/leonelli22a.html
Shenvi, A., Smith, J.Q., Walton, R., Eldridge, S.: Modelling with non-stratified chain event graphs. In: International Conference on Bayesian Statistics in Action, pp. 155–163. Springer (2018)
Silander, T., Leong, T.Y.: A dynamic programming algorithm for learning Chain Event Graphs. In: International Conference on Discovery Science, pp. 201–216. Springer (2013)
Smith, J.Q., Anderson, P.E.: Conditional independence and chain event graphs. Artif. Intell. 172(1), 42–68 (2008)
Strong, P., McAlpine, A., Smith, J.Q.: A Bayesian Analysis of Migration Pathways using Chain Event Graphs of Agent Based Models (2021). ArXiv preprint. arXiv:2111.04368
Strong, P., Smith, J.Q.: Bayesian model averaging of chain event graphs for robust explanatory modelling. In: Salmerrón, A., Rumı, R. (eds.) Proceedings of The 11th International Conference on Probabilistic Graphical Models, Proceedings of Machine Learning Research, vol. 186, pp. 61–72. PMLR (2022). https://proceedings.mlr.press/v186/strong22a.html
Walley, G., Shenvi, A., Strong, P., Kobalczyk, K.: Cegpy: Modelling with Chain Event Graphs in Python (2022). https://doi.org/10.48550/ARXIV.2211.11366, arXiv:2211.11366
Yu, X., Smith, J.Q.: Causal algebras on chain event graphs with informed missingness for system failure. Entropy 23(10) (2021). https://doi.org/10.3390/e23101308, https://www.mdpi.com/1099-4300/23/10/1308
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Strong, P., Smith, J.Q. (2023). Scalable Model Selection for Staged Trees: Mean-posterior Clustering and Binary Trees. In: Avalos-Pacheco, A., De Vito, R., Maire, F. (eds) Bayesian Statistics, New Generations New Approaches. BAYSM 2022. Springer Proceedings in Mathematics & Statistics, vol 435. Springer, Cham. https://doi.org/10.1007/978-3-031-42413-7_3
Download citation
DOI: https://doi.org/10.1007/978-3-031-42413-7_3
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-42412-0
Online ISBN: 978-3-031-42413-7
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)