Model Averaging and Its Use in Economics
The DOI matches its Crossref record; 306 of the 306 references it cites were found in a registry or an index.
Abstract
The method of model averaging has become an important tool to deal with model uncertainty, for example in situations where a large amount of different theories exist, as are common in economics. Model averaging is a natural and formal response to model uncertainty in a Bayesian framework, and most of the paper deals with Bayesian model averaging. The important role of the prior assumptions in these Bayesian procedures is highlighted. In addition, frequentist model averaging methods are also discussed. Numerical techniques to implement these methods are explained, and I point the reader to some freely available computational resources. The main focus is on uncertainty regarding the choice of covariates in normal linear regression models, but the paper also covers other, more challenging, settings, with particular emphasis on sampling models commonly used in economics. Applications of model averaging in economics are reviewed and discussed in a wide range of areas including growth economics, production modeling, finance and forecasting macroeconomic quantities. (JEL C11, C15, C20, C52, O47).
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.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The cited source says what the claim attributes to it.
In the citing paper
“Empirical work in these domain-specific areas is subject to significant uncertainty about model specification. This may be the consequence of the lack of a universally accepted theory that has been empirically verified as a (near) perfect explanation of reality (theory uncertainty), the multiple ways in which theories can be empirically tested (specification uncertainty), heterogeneity uncertainty”
From Modelling Motor Insurance Claim Frequency and Severity Using Gradient Boosting (2023) · cites this paper as Steel 2020
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 · cited 7×
“Comparisons of some methods (including the method by Benjamini and Hochberg (1995) aimed at controlling the false discovery rate) can be found in Deckers and Hanck (2014) in the context of cross-sectional growth regression.”
MethodInfluentialRegression 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 · cited 2×
“ed regression coefficients in the full model will be close to zero. A common classical method is penalized least squares, such as LASSO (least absolute shrinkage and selection operator), introduced by Tibshirani (1996), where the regression “fit” is maximized subject to a complexity penalty. Choosing a different penalty…”
MethodEstimating the Dimension of a Model
Schwarz, G. (1978) · The Annals of Statistics · doi:10.1214/aos/1176344136 · Cited by 47,033 · cited 2×
“The Bayesian information criterion (BIC) implies a stronger complexity penalty and was originally motivated through asymptotic equivalence with a Bayes factor (Schwarz, 1978).”
BackgroundMethodThe Elements of Statistical Learning
Hastie, T., Tibshirani, R., & Friedman, J. (2009) · Springer Series in Statistics · doi:10.1007/978-0-387-84858-7 · Cited by 25,970
The Elements of Statistical Learning
Hastie, T., Friedman, J., & Tibshirani, R. (2001) · Springer Series in Statistics · doi:10.1007/978-0-387-21606-5 · Cited by 24,162
McCullagh, P., & Nelder, J. A. (1989) · doi:10.1007/978-1-4899-3242-6 · Cited by 16,836
Filders, R. (1992) · International Journal of Forecasting · doi:10.1016/0169-2070(92)90130-2 · Cited by 16,165
Kass, R. E., & Raftery, A. E. (1995) · Journal of the American Statistical Association · doi:10.1080/01621459.1995.10476572 · Cited by 12,558
Economic Growth in a Cross Section of Countries
Barro, R. J. (1991) · The Quarterly Journal of Economics · doi:10.2307/2937943 · Cited by 9,263
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 · cited 3×
“Such approaches are e.g. the SCAD penalized regression of Fan and Li (2001) and adaptive LASSO methods as in Zou (2006).”
BackgroundMethod
Research works printed but not linked: 36
- Bernardo, J. and A. Smith (1994). Bayesian Theory . Chichester: Wiley.
- Blażejowski Marcin, 2020, Journal of Applied Econometrics. Published ahead of print
- Blażejowski Marcin, 2015, Journal of Statistical Software 68 (5).
- 10.1002/(SICI)1099-128X(199805/06)12:3<173::AID-CEM505>3.0.CO;2-0 · DOI
- Burnham, K. and D. Anderson (2002). Model selection and multimodel inference: a practical information-theoretic approach (2nd ed.) . New York: Springet.
- Chen, M. and J. Ibrahim (2003). Conjugate priors for generalized linear models. Statistica Sinica 13 , 461-76.
- Clyde, M. and E. Iversen (2013). Bayesian model averaging in the M-open framework. In P. Damien, P. Dellaportas, N. Polson, and D. Stephens (Eds.), Bayesian Theory and Applications , Oxford: Oxford University Press, pp. 483-98.
- Domingos, P. (2000). Bayesian averaging of classifiers and the overfitting problem. In Proceedings of the Seventeenth International Conference on Machine Learning , Stanford, CA, pp. 223-30.
- George, E. (1999a). Comment on 'Bayesian model averaging: A tutorial' by J. Hoeting, D. Madigan, A. Raftery and C. Volinksy. Statistical Science 14 , 409-12.
- George, E. (1999b). Discussion of 'Bayesian model averaging and model search strategies' by M. Clyde. In J. Bernardo, J. Berger, A. Dawid, and A. Smith (Eds.), Bayesian Statistics 6 , Oxford: Oxford University Press, pp. 175-7.
- George, E. and R. McCulloch (1997). Approaches for Bayesian variable selection. Statistica Sinica 7 , 339-73.
- Hanck, C. (2016). I just ran two trillion regressions. Economics Bulletin 36 , 2017-42.
- Henderson, D. and C. Parmeter (2016). Model averaging over nonparametric estimators. In G. Gonz´ alez-Rivera, R. Carter Hill, and T.-H. Lee (Eds.), Essays in Honor of Aman Ullah , Advances in Econometrics, Vol. 36, Emerald Group Publishing Limited, pp. 539-60.
- Jeffreys, H. (1961). Theory of Probability (3rd ed.). Oxford University Press.
- Kapetanios, G. and F. Papailias (2018). Big data & macroeconomic nowcasting: Methodological review. Discussion Paper 2018-12, Economic Statistics Centre of Excellence.
- Koop, G. (2003). Bayesian Econometrics . Chichester: Wiley.
- Kourtellos, A. and C. Tsangarides (2015). Robust correlates of growth spells: Do inequality and redistribution matter? Working Paper 15-20, The Rimini Centre for Economic Analysis.
- Leamer, E. (1978). Specification Searches: Ad Hoc Inference with Nonexperimental Data . New York: Wiley.
- Leamer, E. (1985). Sensitivity analyses would help. American Economic Review 75 , 308-13.
- Li, C., Q. Li, J. Racine, and D. Zhang (2018). Optimal model averaging of varying coefficient models. Statistica Sinica .
- Liang, F. and W. Wong (2000). Evolutionary Monte Carlo: Applications to c p model sampling and change point problem. Statistica Sinica 10 , 317-42.
- Mazerolle Marc J, 2017, Version, 2, 3
- Meng, X. and W. Wong (1996). Simulating ratios of normalizing constants via a simple identity: A theoretical exploration. Statistica Sinica 6 , 831-860.
- Osiewalski, J. and M. Steel (1993). Una perspectiva bayesiana en selecci´ on de modelos. Cuadernos Economicos 55 , 327-51. Original English version 'A Bayesian perspective on model selection' at http://www.cyfronet.krakow.pl/˜eeosiewa/pubo.htm .
- Robert, C. (2007). The Bayesian choice: from decision-theoretic foundations to computational implementation (2nd ed.). Springer.
- Robert, C. and G. Casella (2004). Monte Carlo Statistical Methods . New York: Springer.
- Rossi, P. E., G. Allenby, and R. McCulloch (2006). Bayesian Statistics and Marketing . New York: Wiley.
- Sabanés Bové Daniel, 2011, Bayesian Analysis, 6, 387, 10.1214/ba/1339616469 · doi:10.1214/ba/1339616469
- Sala-i-Martin, X. (1997). I just ran two million regressions. American Economic Review 87 (2), 178-83.
- Shao, J. (1997). An asymptotic theory for linear model selection. Statistica Sinica 7 , 221-64.
- Stock, J. and M. Watson (2006). Forecasting with many predictors. In C. G. G. Elliott and A. Timmermann (Eds.), Handbook of Economic Forecasting , pp. 515-54. Elsevier.
- Volinsky, C., D. Madigan, A. Raftery, and R. Kronmal (1997). Bayesian model averaging in proportional hazards model: Predicting the risk of a stroke. Applied Statistics 46 , 443-8.
- Zellner, A. (1971). An Introduction to Bayesian Inference in Econometrics . New York: Wiley.
- Zellner, A. (1986). On assessing prior distributions and Bayesian regression analysis with g-prior distributions. In P. K. Goel and A. Zellner (Eds.), Bayesian Inference and Decision Techniques: Essays in Honor of Bruno de Finetti , Amsterdam: North-Holland, pp. 233-43.
- Zeugner Stefan, 2014, Bayesian Model Averaging in R. Version, 0, 3
- Zhang, X. and W. Wang (2019). Optimal model averaging estimation for partially linear models. Statistica Sinica , forthcoming.
Cited by
Items 301–323 of 323
Apple Foliar Leaf Disease Detection Using Machine Learning
Kumar, U., Pal, R., & Tomar, S. Y. (2026) · Lecture notes in networks and systems · doi:10.1007/978-3-032-14044-9_25
Model averaging with ridge regularization
Skolkova, A. (2026) · Empirical Economics · doi:10.1007/s00181-025-02866-2
Weighted-Average Least Squares Estimation of Panel Data Models
Luca, G. D., & Magnus, J. R. (2026) · Emerging topics in statistics and biostatistics · doi:10.1007/978-3-031-94050-7_13
Contamination-Robust Frequentist Model Averaging in Predictive RegressionsPreprint
Arvanitis, S., Pinar, M., Stengos, T., & Topaloglou, N. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.6262172
Regime‐Dependent Nowcasting of the Austrian Economy
Hlouskova, J., & Fortin, I. (2026) · Journal of Forecasting · doi:10.1002/for.70123
badp: Bayesian Averaging for Dynamic Panels
Beck, K., Cukier, P., Dubel, M., Szczepanczyk, M., & Wyszyński, M. (2026) · doi:10.32614/cran.package.badp
Malakoff, K. L., Kroetz, K., & Fissel, B. E. (2026) · Journal of Environmental Economics and Management · doi:10.1016/j.jeem.2026.103331
Neural Autoregressive Flows Based Variational Bayes Model Averaging
Zhou, J., & Bhattacharya, S. (2026) · The American Statistician · doi:10.1080/00031305.2026.2654767
“BMA has found widespread application in diverse fields, including economics (Steel 2020), physics (Neil and Sitison 2024), and meta-analysis (Berkhout et al. 2024).”
Raheem, A. A. (2026) · Journal of Public Affairs · doi:10.1002/pa.70137
Brugarolas, P. (2026) · Review of Economics of the Household · doi:10.1007/s11150-026-09847-z · Cited by 2
Bayesian model averaging with non-conjugate priors
Tasiopoulos, A. E., Tsionas, E. G., & Vlastakis, N. D. (2026) · Journal of Econometrics · doi:10.1016/j.jeconom.2026.106256 · Cited by 2
Focused weighted-average least squares estimator
Yin, S.-Y. (2026) · Statistics and Computing · doi:10.1007/s11222-026-10915-8
“As a remedy, a large body of literature has developed around model averaging (MA) methods, whichcombineinformationacross multiplemodels todeliver more robust parameter estimates and predictive performance; see, for example, Steel (2020) for a comprehensive survey.”
Climate risks and forecastability of US inflation: Evidence from dynamic quantile model averaging
Luo, J., Fu, S., Çepni, O., & Gupta, R. (2026) · International Review of Economics & Finance · doi:10.1016/j.iref.2026.105576
“…derived under the mean squared error (MSE) loss or Bayesian frameworks designed mainly for mean regression models (see, for example, the survey papers by Chan and Pauwels (2018) and Steel (2020)), with dynamic forecast combination designed specifically for quantile estimators being a rarity.”
Zou, J., Hu, X., Yan, X., He, B., & Zhang, X. (2026) · Springer eBooks · doi:10.1007/978-981-92-0164-8_1
We are different: the drivers of asset quality in loan type and sectoral breakdowns
Çolak, M. S., Kılıç, Y., Öztürk, H., & Şamcı, M. E. (2026) · Empirica · doi:10.1007/s10663-026-09690-6
“Since FMA considers all contributions out of all possible combinations of independent variables, out of n independent variables, the number of models is equal to 2 n -1, and is constructed as follows: where w i is the weight assigned to each model M i .”
InfluentialAlAskar, I. T., Khatatbeh, I. N., Faras, R., & Bash, A. (2026) · Journal of risk and financial management · doi:10.3390/jrfm19080566
Yuan, C., Li, Y., Fang, F., & Yang, Y. (2026) · Journal of Business and Economic Statistics · doi:10.1080/07350015.2026.2683048
msBMA: Bayesian model averaging over reduced model spacesPreprint
Beck, K. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.7478711
Constrained Bayesian Neural Network utility in the design of price promotions
Burda, M., & Campbell, C. (2026) · Journal of Choice Modelling · doi:10.1016/j.jocm.2026.100633
Wealth and weight revisited: A Bayesian model averaging approach to the Obesity Kuznets Curve
Gómez, K. R., Du, H., & Mendez, C. (2026) · The European Journal of Health Economics · doi:10.1007/s10198-026-01998-1
Digitalisation as a push or pull factor for foreign direct investments: a meta-analysis
Cipollina, M., Pascale, G. D., & Romagno, A. (2026) · Journal of Applied Economics · doi:10.1080/15140326.2026.2739393
Determinants of Hotel Operations under Model Uncertainty
Lin, S., & Chin, K. (2026) · Review of Economics · doi:10.1515/roe-2026-0009
“Essentially, BMA functions as a form of model averaging, mitigating the risks of single-model dependency by formally integrating model uncertainty into the analysis (Moral-Benito 2015; Steel 2020).”
Metadata
- Authors
- Steel, M. F. J.
- Affiliations
- University of Warwick (GB)
- Corresponding author
- Steel, Mark F. J.
- ORCID
- Steel, Mark F. J.
- Venue
- Journal of Economic Literature
- Publisher
- American Economic Association
- Published
- 2020-09-01
- Volume
- 58
- Issue
- 3
- Pages
- 644-719
- ISSN
- 0022-0515 (ISSN-L)
- OpenAlex
- W2759653811
- Cited by
- 321
- References
- 278
- Citations by year
- 2026: 33 · 2025: 52 · 2024: 45 · 2023: 52 · 2022: 47 · 2021: 60 · 2020: 20
- FWCI
- 31.67
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- Journal of Economic Literature (publishedVersion)
- Warwick Research Archive Portal (University of Warwick) (acceptedVersion)
- RePEc: Research Papers in Economics (submittedVersion)
- Topics
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods and Bayesian Inference
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods and Inference
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods and Applications
- Keywords
- model averaging, model uncertainty, Bayesian model averaging, frequentist model averaging, prior assumptions, covariate selection, normal linear regression, sampling model, growth theory, production modeling, macroeconomic forecasting, econometric model selection
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.