Weight Trimming and Propensity Score Weighting
Open accessThe DOI matches its Crossref record; 33 of the 33 references it cites were found in a registry or an index.
Abstract
Propensity score weighting is sensitive to model misspecification and outlying weights that can unduly influence results. The authors investigated whether trimming large weights downward can improve the performance of propensity score weighting and whether the benefits of trimming differ by propensity score estimation method. In a simulation study, the authors examined the performance of weight trimming following logistic regression, classification and regression trees (CART), boosted CART, and random forests to estimate propensity score weights. Results indicate that although misspecified logistic regression propensity score models yield increased bias and standard errors, weight trimming following logistic regression can improve the accuracy and precision of final parameter estimates. In contrast, weight trimming did not improve the performance of boosted CART and random forests. The performance of boosted CART and random forests without weight trimming was similar to the best performance obtainable by weight trimmed logistic regression estimated propensity scores. While trimming may be used to optimize propensity score weights estimated using logistic regression, the optimal level of trimming is difficult to determine. These results indicate that although trimming can improve inferences in some settings, in order to consistently improve the performance of propensity score weighting, analysts should focus on the procedures leading to the generation of weights (i.e., proper specification of the propensity score model) rather than relying on ad-hoc methods such as weight trimming.
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
Breiman, L. (2001) · Machine Learning · doi:10.1023/a:1010933404324 · Cited by 133,538
The central role of the propensity score in observational studies for causal effects
ROSENBAUM, P. R., & RUBIN, D. B. (1983) · Biometrika · doi:10.1093/biomet/70.1.41 · Cited by 31,241
Matching Methods for Causal Inference: A Review and a Look Forward
Stuart, E. A. (2010) · Statistical Science · doi:10.1214/09-sts313 · Cited by 5,478 · cited 1×
“, the weighted or matched samples), ensuring that the groups are comparable with respect to the observed covariates (see Stuart, 2010 [9], or Rubin, 2001 [10], for further discussion of propensity score diagnostics).”
BackgroundSelf-citationConstructing Inverse Probability Weights for Marginal Structural Models
Cole, S. R., & Hernan, M. A. (2008) · American Journal of Epidemiology · doi:10.1093/aje/kwn164 · Cited by 2,796 · cited 3×
“Trimming was performed using percentile cutpoints [15].”
BackgroundVariable Selection for Propensity Score Models
Brookhart, M. A., Schneeweiss, S., Rothman, K. J., Glynn, R. J., Avorn, J., & Stürmer, T. (2006) · American Journal of Epidemiology · doi:10.1093/aje/kwj149 · Cited by 2,336
Using Propensity Scores to Help Design Observational Studies: Application to the Tobacco Litigation
Rubin, D. B. (2001) · Health Services and Outcomes Research Methodology · doi:10.1023/a:1020363010465 · Cited by 2,228 · cited 2×
“, the weighted or matched samples), ensuring that the groups are comparable with respect to the observed covariates (see Stuart, 2010 [9], or Rubin, 2001 [10], for further discussion of propensity score diagnostics).”
BackgroundDoubly robust estimation in missing data and causal inference models
Bang, H., & Robins, J. M. (2005) · Biometrics · doi:10.1111/j.1541-0420.2005.00377.x · Cited by 1,998
McCaffrey, D. F., Ridgeway, G., & Morral, A. R. (2004) · Psychological Methods · doi:10.1037/1082-989x.9.4.403 · Cited by 1,317 · cited 3×
“In particular, the machine learning methods of boosted CART [19] and random forests [25] provided consistently superior performance.”
BackgroundMethodResultAdjusting for Nonignorable Drop-Out Using Semiparametric Nonresponse Models
Scharfstein, D. O., Rotnitzky, A., & Robins, J. M. (1999) · Journal of the American Statistical Association · doi:10.1080/01621459.1999.10473862 · Cited by 995
Improving propensity score weighting using machine learning
Lee, B. K., Lessler, J., & Stuart, E. A. (2010) · Statistics in Medicine · doi:10.1002/sim.3782 · Cited by 896
Self-citation
Cited by
Austin, P. C., & Stuart, E. A. (2015) · Statistics in Medicine · doi:10.1002/sim.6607 · Cited by 4,495
Self-citationA tutorial on propensity score estimation for multiple treatments using generalized boosted models
McCaffrey, D. F., Griffin, B. A., Almirall, D., Slaughter, M. E., Ramchand, R., & Burgette, L. F. (2013) · Statistics in Medicine · doi:10.1002/sim.5753 · Cited by 1,450
“Trimming the weights [19] or removing records with outlier weights improved balance, but it remained inferior to the balance obtained using GBM.”
ResultAustin, P. C. (2014) · Statistics in Medicine · doi:10.1002/sim.5984 · Cited by 1,422
Addressing extreme propensity scores via the overlap weights
Li, F., & Thomas, L. E. (2018) · American Journal of Epidemiology · doi:10.1093/aje/kwy201 · Cited by 694
Desai, R. J., & Franklin, J. M. (2019) · BMJ · doi:10.1136/bmj.l5657 · Cited by 657
Lindahl, B., Baron, T., Erlinge, D., Hadziosmanovic, N., Nordenskjöld, A., Gard, A., & Jernberg, T. (2017) · Circulation · doi:10.1161/circulationaha.116.026336 · Cited by 482
Yao, L., Chu, Z., Li, S., Li, Y., Gao, J., & Zhang, A. (2021) · ACM Transactions on Knowledge Discovery from Data · doi:10.1145/3444944 · Cited by 478
Targeted Maximum Likelihood Estimation for Causal Inference in Observational Studies
Schuler, M. S., & Rose, S. (2017) · American Journal of Epidemiology · doi:10.1093/aje/kww165 · Cited by 399
Adams, R., Henry, K. E., Sridharan, A., Soleimani, H., Zhan, A., Rawat, N., Johnson, L., Hager, D. N., Cosgrove, S. E., Markowski, A., et al. (2022) · Nature Medicine · doi:10.1038/s41591-022-01894-0 · Cited by 365
“We used logistic regression for the probability of treatment model and stabilized and truncated weights at the 1st and 99th weight percentiles to reduce estimation variance 48 .”
MethodAustin, P. C., & Stuart, E. A. (2017) · Statistical Methods in Medical Research · doi:10.1177/0962280215584401 · Cited by 316
Self-citationKost, K., & Lindberg, L. (2015) · Demography · doi:10.1007/s13524-014-0359-9 · Cited by 298
Ali, M. S., Prieto-Alhambra, D., Lopes, L. C., Ramos, D., Bispo, N., Ichihara, M. Y., Pescarini, J. M., Williamson, E., Fiaccone, R. L., Barreto, M. L., et al. (2019) · Frontiers in Pharmacology · doi:10.3389/fphar.2019.00973 · Cited by 200
Contextualizing selection bias in Mendelian randomization: how bad is it likely to be?
Gkatzionis, A., & Burgess, S. (2019) · International Journal of Epidemiology · doi:10.1093/ije/dyy202 · Cited by 182
“It is common in practice to trim weights above some threshold 29 —e.g. to set the largest 1% of weights to be equal to the 99th percentile of the empirical distribution of weights.”
Uncovering survivorship bias in longitudinal mental health surveys during the COVID-19 pandemic
Czeisler, M. É., Wiley, J. F., Czeisler, C. A., Rajaratnam, S. M. W., & Howard, M. E. (2021) · Epidemiology and Psychiatric Sciences · doi:10.1017/s204579602100038x · Cited by 90
“Given the bias-variance compromises associated with trimming survey weights (Lee et al., 2011), no trimming was conducted on the primary analytic sample, which had minimum and maximum weights of 0.”
BackgroundYang, S., & Ding, P. (2018) · Biometrika · doi:10.1093/biomet/asy008 · Cited by 75
Delaney, A., Finnis, M., Bellomo, R., Udy, A., Jones, D., Keijzers, G., MacDonald, S., & Peake, S. (2020) · Emergency Medicine Australasia · doi:10.1111/1742-6723.13394 · Cited by 71
Graham, A. L., Papandonatos, G. D., Erar, B., & Stanton, C. A. (2015) · Health Psychology · doi:10.1037/hea0000278 · Cited by 40
Propensity score weighting: an application to anEarlyHeadStart dental study
Burgette, J. M., Preisser, J. S., & Rozier, R. G. (2016) · Journal of Public Health Dentistry · doi:10.1111/jphd.12106 · Cited by 25
“Several studies have shown improved results with propensity scores derived from GBM compared with LRM, including smaller standard errors for the treatment effects and better covariate balance (5,10,13,14).”
MethodImproving Adherence to Smoking Cessation Treatment: Smoking Outcomes in a Web-based Randomized Trial
Graham, A. L., Papandonatos, G. D., Cha, S., Erar, B., & Amato, M. S. (2018) · Annals of Behavioral Medicine · doi:10.1093/abm/kax023 · Cited by 18
“The final set of weights was inspected for covariate overlap and the need for weight trimming [51] using the balance assessment diagnostics of the TWANG package [52].”
Method(2022) · doi:10.1145/3531146.3533103 · Cited by 1
Metadata
- Authors
- Lee, B. K., Lessler, J., & Stuart, E. A.
- Venue
- PLoS ONE
- Publisher
- Public Library of Science (PLoS)
- Published
- 2011-03-31
- Volume
- 6
- Issue
- 3
- Pages
- e18174
- ISSN
- 1932-6203 (ISSN-L)
- OpenAlex
- W2060316606
- Cited by
- 500
- References
- 33
- FWCI
- 8.90
- Type
- journal article
- Language
- en
- Open access
- gold
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.