Extended Bayesian information criteria for model selection with large model spaces
The DOI matches its Crossref record; 24 of the 24 references it cites were found in a registry or an index.
Abstract
The ordinary Bayesian information criterion is too liberal for model selection when the model space is large. In this paper, we re-examine the Bayesian paradigm for model selection and propose an extended family of Bayesian information criteria, which take into account both the number of unknown parameters and the complexity of the model space. Their consistency is established, in particular allowing the number of covariates to increase to infinity with the sample size. Their performance in various situations is evaluated by simulation studies. It is demonstrated that the extended Bayesian information criteria incur a small loss in the positive selection rate but tightly control the false discovery rate, a desirable property in many applications. The extended Bayesian information criteria are extremely useful for variable selection in problems with a moderate sample size but with a huge number of covariates, especially in genome-wide association studies, which are now an active area in genetics research.
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.
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
Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing
Benjamini, Y., & Hochberg, Y. (1995) · Journal of the Royal Statistical Society Series B: Statistical Methodology · doi:10.1111/j.2517-6161.1995.tb02031.x · Cited by 109,482
Regression shrinkage and selection via the lasso
Tibshirani, R. (1996) · Journal of the Royal Statistical Society Series B: Statistical Methodology · doi:10.1111/j.2517-6161.1996.tb02080.x · Cited by 51,482
Estimating the Dimension of a Model
Schwarz, G. (1978) · The Annals of Statistics · doi:10.1214/aos/1176344136 · Cited by 47,033
Information Theory and an Extension of the Maximum Likelihood Principle
Akaike, H. (1998) · Springer Series in Statistics · doi:10.1007/978-1-4612-1694-0_15 · Cited by 17,781
Variable Selection via Nonconcave Penalized Likelihood and its Oracle Properties
Fan, J., & Li, R. (2001) · Journal of the American Statistical Association · doi:10.1198/016214501753382273 · Cited by 8,900
The Large-Sample Distribution of the Likelihood Ratio for Testing Composite Hypotheses
Wilks, S. S. (1938) · The Annals of Mathematical Statistics · doi:10.1214/aoms/1177732360 · Cited by 3,493
Craven, P., & Wahba, G. (1978) · Numerische Mathematik · doi:10.1007/bf01404567 · Cited by 3,388
High dimensional graphs and variable selection with the Lasso
Meinshausen, N., & Bühlmann, P. (2006) · Cited by 2,539
High-dimensional graphs and variable selection with the Lasso
Meinshausen, N., & Bühlmann, P. (2006) · The Annals of Statistics · doi:10.1214/009053606000000281 · Cited by 2,521 · cited 3×
“Furthermore, unlike competitors such as that of Meinshausen & Bühlmann (2006), the extended Bayes information criterion family does not require a data adaptive tuning parameter procedure in order to be consistent, and hence is easy to use in applications.”
BackgroundLimit theorems in change-point analysis
Csörgő, M., & Horvath, L. H. (1997) · Cited by 1,366
Cited by
A tutorial on regularized partial correlation networks
Epskamp, S., & Fried, E. I. (2018) · Psychological Methods · doi:10.1037/met0000167 · Cited by 2,918
Comparing network structures on three aspects: A permutation test.
van Borkulo, C. D., van Bork, R., Boschloo, L., Kossakowski, J. J., Tio, P., Schoevers, R. A., Borsboom, D., & Waldorp, L. J. (2023) · Psychological Methods · doi:10.1037/met0000476 · Cited by 1,465
The Gaussian Graphical Model in Cross-Sectional and Time-Series Data
Epskamp, S., Waldorp, L. J., Mõttus, R., & Borsboom, D. (2018) · Multivariate Behavioral Research · doi:10.1080/00273171.2018.1454823 · Cited by 1,134
Segura, V., Vilhjálmsson, B. J., Platt, A., Korte, A., Seren, Ü., Long, Q., & Nordborg, M. (2012) · Nature Genetics · doi:10.1038/ng.2314 · Cited by 1,072
Golino, H. F., & Epskamp, S. (2017) · PLOS ONE · doi:10.1371/journal.pone.0174035 · Cited by 1,060
Network analysis: a brief overview and tutorial
Hevey, D. (2018) · Health Psychology and Behavioral Medicine · doi:10.1080/21642850.2018.1521283 · Cited by 1,058
Huang, M., Liu, X., Zhou, Y., Summers, R. M., & Zhang, Z. (2019) · GigaScience · doi:10.1093/gigascience/giy154 · Cited by 855
Association of Symptom Network Structure With the Course of Depression
van Borkulo, C., Boschloo, L., Borsboom, D., Penninx, B. W. J. H., Waldorp, L. J., & Schoevers, R. A. (2015) · JAMA Psychiatry · doi:10.1001/jamapsychiatry.2015.2079 · Cited by 846
A new method for constructing networks from binary data
van Borkulo, C. D., Borsboom, D., Epskamp, S., Blanken, T. F., Boschloo, L., Schoevers, R. A., & Waldorp, L. J. (2014) · Scientific Reports · doi:10.1038/srep05918 · Cited by 757
Fried, E. I., Epskamp, S., Nesse, R. M., Tuerlinckx, F., & Borsboom, D. (2016) · Journal of Affective Disorders · doi:10.1016/j.jad.2015.09.005 · Cited by 751
Sensitivity and specificity of information criteria
(2019) · doi:10.1093/bib/bbz016 · Cited by 718
Advances in surrogate based modeling, feasibility analysis, and optimization: A review
Bhosekar, A., & Ierapetritou, M. (2018) · Computers & Chemical Engineering · doi:10.1016/j.compchemeng.2017.09.017 · Cited by 707
Extended Bayesian Information Criteria for Gaussian Graphical Models
Foygel, R., & Drton, M. (2010) · arXiv (Cornell University) · doi:10.48550/arxiv.1011.6640 · Cited by 659
Variable selection in nonparametric additive models
Huang, J., Horowitz, J. L., & Wei, F. (2010) · The Annals of Statistics · doi:10.1214/09-aos781 · Cited by 537
Golino, H., Shi, D., Christensen, A. P., Garrido, L. E., Nieto, M. D., Sadana, R., Thiyagarajan, J. A., & Martinez-Molina, A. (2020) · Psychological Methods · doi:10.1037/met0000255 · Cited by 532
Multiple predicting K -fold cross-validation for model selection
Jung, Y. (2018) · Journal of Nonparametric Statistics · doi:10.1080/10485252.2017.1404598 · Cited by 499
Generalized Network Psychometrics: Combining Network and Latent Variable Models
Epskamp, S., Rhemtulla, M., & Borsboom, D. (2017) · Psychometrika · doi:10.1007/s11336-017-9557-x · Cited by 471
Network Analysis of Psychopathology: Controversies and Challenges
McNally, R. J. (2021) · Annual Review of Clinical Psychology · doi:10.1146/annurev-clinpsy-081219-092850 · Cited by 463
Model selection may not be a mandatory step for phylogeny reconstruction
Abadi, S., Azouri, D., Pupko, T., & Mayrose, I. (2019) · Nature Communications · doi:10.1038/s41467-019-08822-w · Cited by 406
Model Evaluation of Continuous Data Pharmacometric Models: Metrics and Graphics
Nguyen, T. H. T., Mouksassi, M., Holford, N., Al‐Huniti, N., Freedman, I., Hooker, A. C., John, J., Karlsson, M. O., Mould, D. R., Pérez Ruixo, J. J., et al. (2017) · CPT: Pharmacometrics & Systems Pharmacology · doi:10.1002/psp4.12161 · Cited by 402
Isvoranu, A.-M., & Epskamp, S. (2023) · Psychological Methods · doi:10.1037/met0000439 · Cited by 401
A Network Approach to Psychosis: Pathways Between Childhood Trauma and Psychotic Symptoms
Isvoranu, A.-M., van Borkulo, C. D., Boyette, L.-L., Wigman, J. T. W., Vinkers, C. H., Borsboom, D., & Group Investigators (2017) · Schizophrenia Bulletin · doi:10.1093/schbul/sbw055 · Cited by 397
Alterations in Enteric Virome Are Associated With Colorectal Cancer and Survival Outcomes
Nakatsu, G., Zhou, H., Wu, W. K. K., Wong, S. H., Coker, O. O., Dai, Z., Li, X., Szeto, C.-H., Sugimura, N., Lam, T. Y.-T., et al. (2018) · Gastroenterology · doi:10.1053/j.gastro.2018.04.018 · Cited by 386
Self-citationModel Selection Techniques: An Overview
Ding, J., Tarokh, V., & Yang, Y. (2018) · IEEE Signal Processing Magazine · doi:10.1109/msp.2018.2867638 · Cited by 383
Toward a formalized account of attitudes: The Causal Attitude Network (CAN) model.
Dalege, J., Borsboom, D., van Harreveld, F., van den Berg, H., Conner, M., & van der Maas, H. L. J. (2016) · Psychological Review · doi:10.1037/a0039802 · Cited by 373
Forward Regression for Ultra-High Dimensional Variable Screening
Wang, H. (2009) · Journal of the American Statistical Association · doi:10.1198/jasa.2008.tm08516 · Cited by 366
Xue, H., Shen, X., & Pan, W. (2021) · The American Journal of Human Genetics · doi:10.1016/j.ajhg.2021.05.014 · Cited by 358
Christensen, A. P., & Golino, H. (2021) · Psych · doi:10.3390/psych3030032 · Cited by 339
Estimating psychopathological networks: Be careful what you wish for
Epskamp, S., Kruis, J., & Marsman, M. (2017) · PLOS ONE · doi:10.1371/journal.pone.0179891 · Cited by 315
Outlier Detection Using Nonconvex Penalized Regression
She, Y., & Owen, A. B. (2011) · Journal of the American Statistical Association · doi:10.1198/jasa.2011.tm10390 · Cited by 310
Co-morbid obsessive–compulsive disorder and depression: A bayesian network approach
McNally, R. J., Mair, P., Mugno, B. L., & Riemann, B. C. (2017) · Psychological Medicine · doi:10.1017/s0033291716003287 · Cited by 308
“Epskamp et al.’s (2012) qgraph package furnishes an extended Bayesian information criterion (EBIC) model comparison procedure that ascertains the tuning parameter that optimizes model fit as well as parsimony (Chen & Chen, 2008), given a specific value of the hyperparameter γ.”
MethodCostantini, G., Richetin, J., Preti, E., Casini, E., Epskamp, S., & Perugini, M. (2019) · Personality and Individual Differences · doi:10.1016/j.paid.2017.06.011 · Cited by 291
The Network Structure of Symptoms of the Diagnostic and Statistical Manual of Mental Disorders
Boschloo, L., van Borkulo, C. D., Rhemtulla, M., Keyes, K. M., Borsboom, D., & Schoevers, R. A. (2015) · PLOS ONE · doi:10.1371/journal.pone.0137621 · Cited by 285
McNally, R. J., Heeren, A., & Robinaugh, D. J. (2017) · European Journal of Psychotraumatology · doi:10.1080/20008198.2017.1341276 · Cited by 272
Körber, V., Yang, J., Barah, P., Wu, Y., Stichel, D., Gu, Z., Fletcher, M. N. C., Jones, D., Hentschel, B., Lamszus, K., et al. (2019) · Cancer Cell · doi:10.1016/j.ccell.2019.02.007 · Cited by 260
Li, Z., & Sillanpää, M. J. (2012) · Theoretical and Applied Genetics · doi:10.1007/s00122-012-1892-9 · Cited by 258
Tuning Parameter Selection in High Dimensional Penalized Likelihood
Fan, Y., & Tang, C. Y. (2013) · Journal of the Royal Statistical Society Series B: Statistical Methodology · doi:10.1111/rssb.12001 · Cited by 256
Epskamp, S., Maris, G., Waldorp, L. J., & Borsboom, D. (2018) · The Wiley Handbook of Psychometric Testing · doi:10.1002/9781118489772.ch30 · Cited by 256
On the equivalency of factor and network loadings
Christensen, A. P., & Golino, H. (2021) · Behavior Research Methods · doi:10.3758/s13428-020-01500-6 · Cited by 253
Lassopack: model selection and prediction with regularized regression in Stata
Ahrens, A., Hansen, C. B., & Schaffer, M. E. (2020) · The Stata Journal Promoting communications on statistics and Stata · doi:10.1177/1536867x20909697 · Cited by 212
“To see why, consider the case where p = 1000 (following Chen and Chen 2008): There are 1000 models for which one parameter is non-zero (s = 1), while there are 1000× 999/2 models for which s = 2.”
InfluentialJackson, C. A., Castro, D. M., Saldi, G.-A., Bonneau, R., & Gresham, D. (2020) · eLife · doi:10.7554/elife.51254 · Cited by 174
Estimating psychological networks and their accuracy: A tutorial paper
Epskamp, S., Borsboom, D., & Fried, E. I. (2018) · Behavior Research Methods · doi:10.3758/s13428-017-0862-1 · Cited by 5,715
Investigating the feasibility of idiographic network models.
Mansueto, A. C., Wiers, R. W., van Weert, J. C. M., Schouten, B. C., & Epskamp, S. (2023) · Psychological Methods · doi:10.1037/met0000466 · Cited by 147
Solmi, M., Collantoni, E., Meneguzzo, P., Degortes, D., Tenconi, E., & Favaro, A. (2018) · International Journal of Eating Disorders · doi:10.1002/eat.22884 · Cited by 119
Bekhuis, E., Schoevers, R. A., van Borkulo, C. D., Rosmalen, J. G. M., & Boschloo, L. (2016) · Psychological Medicine · doi:10.1017/s0033291716001550 · Cited by 89
“To find the best-fitting model, the extended Bayesian information criterion is used to assess goodness of fit (Chen & Chen, 2008).”
MethodGámez-Guadix, M., Sorrel, M. A., & Martínez-Bacaicoa, J. (2023) · Sexuality Research and Social Policy · doi:10.1007/s13178-022-00775-y · Cited by 65
“This procedure requires establishing a penalty parameter (i.e., the extended Bayesian information criterion; Chen & Chen, 2008).”
MethodCaught in a web of trauma: Network analysis of childhood adversity and adult mental ill-health
Breuer, F., Greggersen, W., Kahl, K. G., Schweiger, U., & Westermair, A. L. (2020) · Child Abuse & Neglect · doi:10.1016/j.chiabu.2020.104534 · Cited by 47
Zeiler, M., Philipp, J., Truttmann, S., Waldherr, K., Wagner, G., & Karwautz, A. (2021) · Nutrients · doi:10.3390/nu13114096 · Cited by 41
“As gLASSO produces a collection of network solution, the Extended Bayesian Information Criterion (EBIC, [46]) was used to select the optimal network model.”
MethodSaba, V., Premi, E., Cristillo, V., Gazzina, S., Palluzzi, F., Zanetti, O., Gasparotti, R., Padovani, A., Borroni, B., & Grassi, M. (2019) · Frontiers in Neuroscience · doi:10.3389/fnins.2019.00211 · Cited by 33
“We further applied two methods defining per subject optimal correlation thresholds, based on the extended Bayesian information criterion (EBIC; Chen and Chen, 2008), and spectral analysis (Perkins and Langston, 2009).”
MethodRen, L., Yang, Z., Wang, Y., Cui, L.-B., Jin, Y., Ma, Z., Zhang, Q., Wu, Z., Wang, H.-N., & Yang, Q. (2020) · BMC Psychiatry · doi:10.1186/s12888-020-02849-w · Cited by 33
“The qgraph package provides an extended Bayesian information criterion (EBIC) to identify the tuning parameter that optimizes the fit and parsimony of the model, and it gives a specific hyperparameter gamma value [75].”
MethodAnxiety and Depression Symptoms among Youth Survivors of Childhood Sexual Abuse: A Network Analysis
Li, J., Jin, Y., Xu, S., Luo, X., Wilson, A., Li, H., Wang, X., Sun, X., & Wang, Y. (2023) · BMC Psychology · doi:10.1186/s40359-023-01275-3 · Cited by 29
“Mean-while, the shrinkage parameter could be selected based on the Extended Bayesian Information Criterion (EBIC) [52].”
MethodPeters, J. R., Crowe, M. L., Morgan, T., Zimmerman, M., Sharp, C., Grilo, C. M., Sanislow, C. A., Shea, M. T., Zanarini, M. C., McGlashan, T. H., et al. (2023) · Psychological Medicine · doi:10.1017/s0033291721004931 · Cited by 27
“Networks were estimated individually for each sample using an Extended Bayesian Information Criterion (Chen & Chen, 2008) (EBIC) graphical LASSO (Tibshirani, 1996) regularization.”
Ruiz, F. J., Bianchi, J. M., Bastidas-Suárez, D. M., Ramírez, E. S., & Peña-Hernández, V. (2024) · Journal of Contextual Behavioral Science · doi:10.1016/j.jcbs.2024.100854 · Cited by 33
A Network Analysis Approach on the Psychological Flexibility/Inflexibility Model
Christodoulou, A., Karekla, M., Costantini, G., & Michaelides, M. P. (2023) · Behavior Therapy · doi:10.1016/j.beth.2023.01.002 · Cited by 22
Multiple paths to rumination within a network analytical framework
Tamm, G., Koster, E. H. W., & Hoorelbeke, K. (2024) · Scientific Reports · doi:10.1038/s41598-024-61469-6 · Cited by 22
Parker, S. F., Rogge, R. D., Drake, C. E., & Fogle, C. (2024) · Journal of Contextual Behavioral Science · doi:10.1016/j.jcbs.2024.100753 · Cited by 22
Yang, T., He, Y., Wu, L., Ren, L., Lin, J., Wang, C., Wu, S., & Liu, X. (2023) · Heliyon · doi:10.1016/j.heliyon.2023.e20938 · Cited by 21
“The extended Bayesian information criterion (EBIC) [33] and least absolute shrinkage and selection operator (LASSO) [34] regularization were applied jointly to punish the trivial edges exactly to zero [35].”
Navarrete, J., Colomer-Carbonell, A., Sanabria-Mazo, J. P., Luciano, J. V., Soler, J., García-Campayo, J., Demarzo, M., Montero-Marín, J., Bohlmeijer, E. T., Campos, D., et al. (2023) · Journal of Contextual Behavioral Science · doi:10.1016/j.jcbs.2023.05.001 · Cited by 5
Metadata
- Authors
- Chen, J., & Chen, Z.
- Venue
- Biometrika
- Publisher
- Oxford University Press (OUP)
- Published
- 2008-09-01
- Volume
- 95
- Issue
- 3
- Pages
- 759-771
- ISSN
- 0006-3444 (ISSN-L)
- 1464-3510
- OpenAlex
- W2053061982
- Cited by
- 2,011
- References
- 24
- FWCI
- 19.25
- Type
- journal article
- Language
- en
- Open access
- closed
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.