Marginalized multilevel models and likelihood inference (with comments and a rejoinder by the authors)
Open accessThe DOI matches its Crossref record; 70 of the 70 references it cites were found in a registry or an index.
Abstract
Hierarchical or ‘‘multilevel’’ regression models typically parameterize the mean response conditional on unobserved latent variables or ‘‘random’’ effects and then make simple assumptions regarding their distribution. The interpretation of a regression parameter in such a model is the change in possibly transformed mean response per unit change in a particular predictor having controlled for all conditioning variables including the random effects. An often overlooked limitation of the conditional formulation for nonlinear models is that the interpretation of regression coefficients and their estimates can be highly sensitive to difficult-to-verify assumptions about the distribution of random effects, particularly the dependence of the latent variable distribution on covariates. In this article, we present an alternative parameterization for the multilevel model in which the marginal mean, rather than the conditional mean given random effects, is regressed on covariates. The impact of random effects model violations on the marginal and more traditional conditional parameters is compared through calculation of asymptotic relative biases. A simple two-level example from a study of teratogenicity is presented where the binomial overdispersion depends on the binary treatment assignment and greatly influences likelihood-based estimates of the treatment effect in the conditional model. A second example considers a three-level structure where attitudes toward abortion over time are correlated with person and district level covariates. We observe that regression parameters in conditionally specified models are more sensitive to random effects assumptions than their counterparts in the marginal formulation.
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
(2005) · Statistics for Biology and Health · doi:10.1007/0-387-27255-0_9 · Cited by 20,133
Longitudinal data analysis using generalized linear models
LIANG, K.-Y., & ZEGER, S. L. (1986) · Biometrika · doi:10.1093/biomet/73.1.13 · Cited by 18,550
Self-citation(1972) · Cited by 17,806
McCullagh, P., & Nelder, J. A. (1989) · doi:10.1007/978-1-4899-3242-6 · Cited by 16,836
Markov Chain Monte Carlo in Practice
(1995) · Chapman & Hall/CRC eBooks · doi:10.1201/b14835 · Cited by 7,757
(2001) · Statistics for Psychologists · doi:10.4324/9781410605542-11 · Cited by 6,877
Statistics and Causal Inference
Holland, P. W. (1986) · Journal of the American Statistical Association · doi:10.1080/01621459.1986.10478354 · Cited by 5,293 · cited 2×
“Second, the multilevel regression parameter 3jC has a desirable causal interpretation (Holland, 1986) as the change in (possibly transformed) expected response per unit change in Xj, holding the other observed variables and unobserved latent factors fixed.”
Background(2006) · Technometrics · doi:10.1198/tech.2006.s417 · Cited by 4,950
Approximate Inference in Generalized Linear Mixed Models
Breslow, N. E., & Clayton, D. G. (1993) · Journal of the American Statistical Association · doi:10.1080/01621459.1993.10594284 · Cited by 4,629
Models for Longitudinal Data: A Generalized Estimating Equation Approach
Zeger, S. L., Liang, K.-Y., & Albert, P. S. (1988) · Biometrics · doi:10.2307/2531734 · Cited by 4,505
Self-citation
Cited by
The R Package geepack for Generalized Estimating Equations
Halekoh, U., Højsgaard, S., & Yan, J. (2006) · Journal of Statistical Software · doi:10.18637/jss.v015.i02 · Cited by 1,907
Hubbard (2010) · doi:10.1097/ede.0b013e3181caeb90 · Cited by 1,139
Income Inequality and Health: What Have We Learned So Far?
Subramanian, S. V. (2004) · Epidemiologic Reviews · doi:10.1093/epirev/mxh003 · Cited by 900
Saltelli, A., & Tarantola, S. (2002) · Journal of the American Statistical Association · doi:10.1198/016214502388618447 · Cited by 447
Hipp, J. R. (2007) · American Sociological Review · doi:10.1177/000312240707200501 · Cited by 414
Ghisletta, P., & Spini, D. (2004) · Journal of Educational and Behavioral Statistics · doi:10.3102/10769986029004421 · Cited by 388
Multilevel modelling of medical data
Goldstein, H., Browne, W., & Rasbash, J. (2002) · Statistics in Medicine · doi:10.1002/sim.1264 · Cited by 388
A tutorial on count regression and zero-altered count models for longitudinal substance use data.
Atkins, D. C., Baldwin, S. A., Zheng, C., Gallop, R. J., & Neighbors, C. (2013) · Psychology of Addictive Behaviors · doi:10.1037/a0029508 · Cited by 387
Maximum Likelihood Estimation in Semiparametric Regression Models with Censored Data
Zeng, D., & Lin, D. Y. (2007) · Journal of the Royal Statistical Society Series B: Statistical Methodology · doi:10.1111/j.1369-7412.2007.00606.x · Cited by 315
Misspecified maximum likelihood estimates and generalised linear mixed models
Heagerty, P. J., & Kurland, B. F. (2001) · Biometrika · doi:10.1093/biomet/88.4.973 · Cited by 294
Self-citationRodríguez, A., Dunson, D. B., & Gelfand, A. E. (2008) · Journal of the American Statistical Association · doi:10.1198/016214508000000553 · Cited by 294
Bayesian Multilevel Analysis and MCMC
Draper, D. (2007) · Handbook of Multilevel Analysis · doi:10.1007/978-0-387-73186-5_2 · Cited by 69
“I mentioned in Section 2.3.2 that the optimal acceptance rate for one-dimensional (random-walk) MH samplers with Gaussian PDs when the target distribution is approximately normal is about 44%; Browne and Draper [11, 12, 13] used this fact to equip MLwiN with the following simple adaptive method for choosing the PD var…”
BackgroundInfluentialRaudenbush, S. W. (2007) · Handbook of Multilevel Analysis · doi:10.1007/978-0-387-73186-5_5 · Cited by 51
“leading Heagerty and Zeger [15] to recommend population-average inference with robust standard errors as a robust alternative to “unit-specific” models of the type reflected in (5.”
BackgroundMethodStepped wedge cluster randomized trials: a methodological overview
Li, F., & Wang, R. (2022) · World Neurosurgery · doi:10.1016/j.wneu.2021.10.136 · Cited by 37
Center-Based Care in the Context of One-Child Policy in China: Do Child Gender and Siblings Matter?
Zhai, F., & Gao, Q. (2010) · Population Research and Policy Review · doi:10.1007/s11113-009-9171-4 · Cited by 17
Metadata
- Authors
- Heagerty, P. J., & Zeger, S. L.
- Venue
- Statistical Science
- Publisher
- Institute of Mathematical Statistics
- Published
- 2000-02-01
- Volume
- 15
- Issue
- 1
- Pages
- 1-26
- ISSN
- 0883-4237 (ISSN-L)
- OpenAlex
- W1535258181
- Cited by
- 298
- References
- 70
- FWCI
- 6.00
- Type
- journal article
- Language
- en
- Open access
- bronze
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.