Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat
Cited by 205

The DOI matches its Crossref record; 11 of the 11 references it cites were found in a registry or an index.

Back to the basics: Rethinking partial correlation network methodology

Open access
Williams, Donald R., Rast, Philippe · 2020 · British Journal of Mathematical and Statistical Psychology

Abstract

The Gaussian graphical model (GGM) is an increasingly popular technique used in psychology to characterize relationships among observed variables. These relationships are represented as elements in the precision matrix. Standardizing the precision matrix and reversing the sign yields corresponding partial correlations that imply pairwise dependencies in which the effects of all other variables have been controlled for. The graphical lasso (glasso) has emerged as the default estimation method, which uses ℓ 1 ‐based regularization. The glasso was developed and optimized for high‐dimensional settings where the number of variables ( p ) exceeds the number of observations ( n ), which is uncommon in psychological applications. Here we propose to go ‘back to the basics’, wherein the precision matrix is first estimated with non‐regularized maximum likelihood and then Fisher Z transformed confidence intervals are used to determine non‐zero relationships. We first show the exact correspondence between the confidence level and specificity, which is due to 1 minus specificity denoting the false positive rate (i.e., α). With simulations in low‐dimensional settings ( p ≪ n ), we then demonstrate superior performance compared to the glasso for detecting the non‐zero effects. Further, our results indicate that the glasso is inconsistent for the purpose of model selection and does not control the false discovery rate, whereas the proposed method converges on the true model and directly controls error rates. We end by discussing implications for estimating GGMs in psychology.

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.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1111/bmsp.12173

Cites

11 of 11 references verified0 citation errors · 0 not checkable

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.

A few free checks a day · papers already verified open instantly · results are public

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

Show all 11 references

Research works printed but not linked: 60

  1. 10.1093/biomet/asr054 · DOI · Paper page
  2. 10.1146/annurev-statistics-022513-115545 · DOI · Paper page
  3. 10.1214/08-AOS606 · DOI
  4. 10.1198/jasa.2011.tm10159 · DOI · Paper page
  5. 10.1016/j.jrp.2014.07.003 · DOI · Paper page
  6. 10.3150/15-BEJ756 · DOI · Paper page
  7. 10.2307/2528966 · DOI · Paper page
  8. 10.1016/j.schres.2019.01.046 · DOI
  9. 10.1080/01621459.2013.823775 · DOI · Paper page
  10. Epskamp S.(2016).Regularized Gaussian psychological networks: Brief report on the performance of extended BIC model selection.arXiv 1606:05771v1.
  11. 10.18637/jss.v048.i04 · DOI · Paper page
  12. 10.1371/journal.pone.0179891 · DOI · Paper page
  13. 10.1080/00273171.2018.1454823 · DOI · Paper page
  14. 10.1198/016214501753382273 · DOI · Paper page
  15. 10.1111/ectj.12061 · DOI · Paper page
  16. Fisher R. A., 1915, Frequency distribution of the values of the correlation coefficient in samples from an indefinitely large population, Biometrika, 10, 507
  17. Fisher R. A., 1924, The distribution of the partial correlation coefficient, Metron, 3, 329
  18. 10.1037/abn0000276 · DOI · Paper page
  19. Foygel R., 2010, Extended Bayesian information criteria for Gaussian graphical models, Advances in Neural Information Processing Systems, 604
  20. 10.1177/2167702617745092 · DOI · Paper page
  21. 10.1093/biostatistics/kxm045 · DOI · Paper page
  22. Gao X. Pu D. Q. Wu Y. &Xu H.(2009).Tuning parameter selection for penalized likelihood estimation of inverse covariance matrix.arXiv.https://doi.org/10.5705/ss.2009.210 · doi:10.5705/ss.2009.210
  23. 10.1111/biom.12186 · DOI
  24. 10.1051/0004-6361:20066170 · DOI · Paper page
  25. 10.1201/b18401 · DOI · Paper page
  26. 10.1214/15-EJS1031 · DOI
  27. 10.1007/s11749-016-0503-5 · DOI
  28. Javanmard A. &Montanari A.(2015).De‐biasing the Lasso: Optimal sample size for Gaussian designs.arXiv.
  29. 10.1371/journal.pone.0148171 · DOI
  30. 10.1002/wics.1415 · DOI
  31. 10.3905/jpm.2004.110 · DOI · Paper page
  32. 10.1214/15-AOS1371 · DOI · Paper page
  33. Leng C., 2006, A note on the lasso and related procedures in model selection, Statistica Sinica, 16, 1273
  34. 10.1016/j.ijar.2017.01.001 · DOI
  35. 10.1080/10618600.2019.1575744 · DOI
  36. 10.1214/13-AOS1169 · DOI · Paper page
  37. Liu H. Roeder K. &Wasserman L.(2010).Stability approach to regularization selection (StARS) for High dimensional graphical models.arXiv 1–14.
  38. 10.1146/annurev.psych.51.1.201 · DOI · Paper page
  39. Mazumder R., 2012, Exact covariance thresholding into connected components for large‐scale graphical lasso, Journal of Machine Learning Research, 13, 781
  40. 10.1080/00273171.2015.1036965 · DOI · Paper page
  41. 10.1214/14-BA889 · DOI
  42. Norouzi M. Fleet D. J. &Salakhutdinov R. R.(2012).Hamming distance metric learning. InAdvances in neural information processing systems Lake Tahoe NV pp.1061–1069.
  43. 10.1198/jasa.2009.0126 · DOI · Paper page
  44. 10.1214/14-AOS1286 · DOI · Paper page
  45. 10.1016/j.drugalcdep.2016.02.005 · DOI · Paper page
  46. 10.1093/bioinformatics/bti062 · DOI · Paper page
  47. 10.2202/1544-6115.1175 · DOI · Paper page
  48. Shao J., 1997, An asymptotic theory for linear model selection, Statistica Sinica, 7, 221
  49. 10.1037/abn0000308 · DOI
  50. 10.1002/cjs.11313 · DOI · Paper page
  51. Tibshirani R., 1996, Regression selection and shrinkage via the lasso, Journal of the Royal Statistical Society Series B, 58, 267, 10.1111/j.2517-6161.1996.tb02080.x · DOI · Paper page
  52. Van Borkulo C. D., 2014, A new method for constructing networks from binary data, Scientific Reports, 4, 1, 10.1038/srep05918 · DOI · Paper page
  53. 10.1371/journal.pcbi.1004755 · DOI
  54. Whittaker J., 1990, Graphical models in applied multivariate statistics
  55. 10.1098/rspa.1907.0028 · DOI
  56. Zhao T., 2012, The huge package for high‐dimensional undirected graph estimation in R, Journal of Machine Learning Research, 13, 1059
  57. Zhao S. Shojaie A. &Witten D.(2017).In defense of the indefensible: A very naive approach to high‐dimensional inference.arXiv(1) 1–61.
  58. Zhao P., 2006, On model selection consistency of lasso, Journal of Machine Learning Research, 7, 2541
  59. 10.1371/journal.pcbi.1006369 · DOI
  60. 10.1089/brain.2017.0511 · DOI

Cited by

Metadata

Authors
Williams, D. R., & Rast, P.
Venue
British Journal of Mathematical and Statistical Psychology
Publisher
Wiley
Published
2019-06-17
Volume
73
Issue
2
Pages
187-212
ISSN
  • 0007-1102 (ISSN-L)
  • 2044-8317
OpenAlex
W2900770950
Cited by
205
References
11
FWCI
18.59
Type
journal article
Language
en
Open access
green
Funders
  • National Academies of Sciences, Engineering, and Medicine FORD foundation pre-doctoral fellowship
  • National Science Foundation Graduate Research Fellowship
  • National Institute on Aging of the National Institutes of Health (R01AG050720)

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.