Abstract
We provide an overview of frequentist model averaging. For point estimation, we consider different methods for selecting the model weights, including those based on AIC, bagging, weighted AIC, stacking and focussed methods. For interval estimation, we consider Wald, MATA and percentile-bootstrap intervals. Use of the methods are illustrated by examples involving real data.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
This can come as a surprise; see [159] for a useful discussion of the assumptions underlying AIC.
- 2.
As discussed in Sect. 2.2, when counting the number of parameters in a model we include any scale parameters, such as the error variance in a normal linear model.
- 3.
See [24] for a discussion of the connection between the model-selection probabilities \(p \left( S = m \right) \) \((m=1,\dots ,M)\) and AIC weights.
- 4.
Throughout the rest of the chapter, it will be implicit that constrained-optimisation is used whenever we determine the weights by minimising an objective function.
- 5.
- 6.
As with AIC(w), the choice of estimate of any scale parameter will not affect the weights.
- 7.
- 8.
- 9.
This assumption has also been used in interval estimation (Sect. 3.4.1).
- 10.
An alternative derivation avoids the notion of selecting a random sample from a population of models [24]. However this involves regarding \(\theta \) as a weighted mean of least-false values of \(\theta \).
- 11.
- 12.
Even if \(\widehat{b}_m\) is unbiased, \(\widehat{b}_m^{\,2}\) will be biased as an estimate of \(b_m^{2}\), but analytical bias-adjustment would involve estimation of the correlation between \(\widehat{\theta }_{m_1}\) and \(\widehat{\theta }_{m_2}\) \((m_1 \ne m_2)\), and any decrease in bias might be offset by an increase in variance.
- 13.
- 14.
- 15.
- 16.
It has been wrongly claimed that use of this interval involves assuming that the largest model is not in the model set [48].
- 17.
- 18.
A similar issue arises when using a technique such as RJMCMC in the Bayesian setting (Sect. 2.2.1), where a large number of iterations may be required in order to visit each model often enough to obtain reliable estimates of both the posterior model probabilities and the posteriors for those parameters in models with low posterior model probabilities.
- 19.
This example also provides evidence that the Wald interval can perform well, despite the issues raised in Sect. 3.4.1.
- 20.
Unless we use DIC weights, which can depend on the paramtetrisation (Sect. 2.5).
- 21.
Conversely, we could adjust the nominal confidence level for each interval until they all have the same width, and then choose the one with the highest true coverage rate [150].
- 22.
- 23.
This constraint can also be useful for generalisation of the conclusions [15].
References
Aiolfi, M., Capistran, C., Timmermann, A.: Forecast combinations. In: Clements, M.P., Hendry, D.F. (eds.) Oxford Handbook of Economic Forecasting. Oxford University Press (2010)
Akaike, H.: Information theory as an extension of the maximum likelihood principle. In: Petrov, B.N., Csaki, F. (eds.) Second International Symposium on Information Theory, pp. 267–281. Akademiai Kiado, Budapest (1973)
Akaike, H.: A new look at the statistical model identification. IEEE Trans. Autom. Control. 19, 716–723 (1974)
Akaike, H.: A Bayesian analysis of the minimum AIC procedure. Ann. I. Stat. Math. 30, 9–14 (1978)
Akaike, H.: A Bayesian extension of the minimum AIC procedure of autoregressive model fitting. Biometrika 66, 237–242 (1979)
Aksu, C., Gunter, S.I.: An empirical analysis of the accuracy of SA, OLS, ERLS and NRLS combination forecasts. Int. J. Forecast. 8, 27–43 (1992)
Amemiya, T.: Selection of regressors. Int. Econ. Rev. 21, 331–354 (1980)
Amini, S.M., Parmeter, C.F.: Comparisons of model averaging techniques: assessing growth determinants. J. Appl. Econ. 27, 870–876 (2012)
Anderson, D.R., Burnham, K.P., White, G.C.: AIC model selection in overdispersed capture-recapture data. Ecology 75, 1780–1793 (1994)
Ando, T., Li, K.-C.: A model-averaging approach for high-dimensional regression. J. Am. Stat. Assoc. 109, 254–265 (2014)
Ando, T., Li, K.-C.: A weight-relaxed model averaging approach for high-dimensional generalized linear models. Ann. Stat. 45, 2654–2679 (2017)
Arlot, S., Celisse, A.: A survey of cross-validation procedures for model selection. Stat. Surv. 4, 40–79 (2010)
Augustin, N., Sauerbrei, W., Schumacher, M.: The practical utility of incorporating model selection uncertainty into prognostic models for survival data. Stat. Model. 5, 95–118 (2005)
Bozdogan, H.: Akaike’s information criterion and recent developments in information complexity. J. Math. Psychol. 44, 62–91 (2000)
Breiman, L.: Stacked regressions. Mach. Learn. 24, 49–64 (1996)
Breiman, L.: Bagging predictors. Mach. Learn. 24, 123–140 (1996)
Breiman, L.: Heuristics of instability and stabilization in model selection. Ann. Stat. 24, 2350–2383 (1996)
Breiman, L.: Random forests. Mach. Learn. 45, 5–32 (2001)
Brewer, M.J., Butler, A., Cooksley, S.L.: The relative performance of AIC, AICC and BIC in the presence of unobserved heterogeneity. Methods Ecol. Evol. 7, 679–692 (2016)
Buchholz, A., Hollnder, N., Sauerbrei, W.: On properties of predictors derived with a two-step bootstrap model averaging approach—a simulation study in the linear regression model. Comput. Stat. Data Anal. 52, 2778–2793 (2008)
Buckland, S.T., Burnham, K.P., Augustin, N.H.: Model selection: an integral part of inference. Biometrics 53, 603–618 (1997)
Buckland, S.T., Burnham, K.P., Augustin, N.H.: Rejoinder to the Letter to the Editors from Wagenmakers, E.-J., Farrell, S., Ratcliff, R. Biometrics 60, 283 (2004)
Burnham, K.P., Anderson, D.R., White, G.C.: Evaluation of the Kullback-Leibler discrepancy for model selection in open population capture-recapture models. Biometrical. J. 36, 299–315 (1994)
Burnham, K.P., Anderson, D.R.: Model Selection and Multimodel Inference: A Practical Information-Theoretic Approach, 2nd edn. Springer (2002)
Burnham, K.P., Anderson, D.R.: Multimodel inference understanding AIC and BIC in model selection. Sociol. Method. Res. 33, 261–304 (2004)
Cade, B.S.: Model averaging and muddled multimodel inferences. Ecology 96, 2370–2382 (2015)
Candolo, C., Davison, A.C., Demtrio, C.G.B.: A note on model uncertainty in linear regression. J. R. Stat. Soc. D-Stat. 52, 165–177 (2003)
Carney, M., Cunningham, P.: Calibrating probability density forecasts with multi-objective search. Technical Report TCD-CS-2006-07, Trinity College, Dublin (2006)
Cavanaugh, J.E.: Unifying the derivations for the Akaike and corrected Akaike information criteria. Stat. Probabil. Lett. 33, 201–208 (1997)
Cavanaugh, J.E., Shumway, R.H.: A bootstrap variant of AIC for state-space model selection. Stat. Sin. 7, 473–496 (1997)
Cavanaugh, J.E.: A large-sample model selection criterion based on Kullback’s symmetric divergence. Stat. Probab. Lett. 42, 333–343 (1999)
Charkhi, A., Claeskens, G., Hansen, B.E.: Minimum mean squared error model averaging in likelihood models. Stat. Sin. 26, 809–840 (2016)
Chen, X., Zou, G., Zhang, X.: Frequentist model averaging for linear mixed-effects models. Front. Math. China 8, 497–515 (2013)
Chung, H.-Y., Lee, K.-W., Koo, J.-Y.: A note on bootstrap model selection criterion. Stat. Probab. Lett. 26, 35–41 (1996)
Claeskens, G., Hjort, N.L.: The focused information criterion. J. Am. Stat. Assoc. 98, 900–916 (2003)
Claeskens, G., Croux, C., Kerckhoven, J.V.: Variable selection for logistic regression using a prediction-focused information criterion. Biometrics 62, 972–979 (2006)
Claeskens, G., Carroll, R.J.: An asymptotic theory for model selection inference in general semiparametric problems. Biometrika 94, 249–265 (2007)
Claeskens, G., Hjort, N.L.: Model Selection and Model Averaging. Cambridge University Press, Cambridge (2008)
Claeskens, G., Magnus, J.R., Vasnev, A.L., Wang, W.: The forecast combination puzzle: a simple theoretical explanation. J. Forecast. 32, 754–762 (2016)
Clyde, M.: Model uncertainty and health effect studies for particulate matter. Environmetrics 11, 745–763 (2000)
Craven, P., Wahba, G.: Smoothing noisy data with spline functions: estimating the correct degree of smoothing by the method of generalized cross-validation. Numer. Math. 31, 377–403 (1979)
Dardanoni, V., Modica, S., Peracchi, F.: Regression with imputed covariates: a generalized missing-indicator approach. J. Econ. 162, 362–368 (2011)
Dardanoni, V., de Luca, G., Modica, S., Peracchi, F.: Bayesian model averaging for generalized linear models with missing covariates. No. 1311. Einaudi Institute for Economics and Finance (2013)
Davison, A.C., Hinkley, D.V.: Bootstrap Methods and their Applications. Cambridge University Press, Cambridge (1997)
Debray, T.P.A., Koffijberg, H., Nieboer, D., Vergouwe, Y., Steyerbergb, E.W., Moonsa, K.G.M.: Meta-analysis and aggregation of multiple published prediction models. Stat. Med. 33, 2341–2362 (2014)
De Luca, G., Magnus, J.R., Peracchi, F.: Weighted-average least squares estimation of generalized linear models. J. Econ. (2018). https://doi.org/10.1016/j.jeconom.2017.12.007
Donohue, M.C., Overholser, R., Xu, R., Vaida, F.: Conditional Akaike information under generalized linear and proportional hazards mixed models. Biometrika 98, 685–700 (2011)
Dormann, C.F., Calabrese, J.M., GuilleraArroita, G., Matechou, E., Bahn, V., Bartoń, K., Beale, C.M., Ciuti, S., Elith, J., Gerstner, K., Guelat, J., Keil, P., LahozMonfort, J.J., Pollock, L.J., Reineking, B., Roberts, D.R., Schröder, B., Thuiller, W., Warton, D.I., Wintle, B.A., Wood, S.N., Wüest, R.O., Hartig, F.: Model averaging in ecology: a review of Bayesian, information-theoretic, and tactical approaches for predictive inference. Ecol. Monogr. (2018). https://doi.org/10.1002/ecm.1309
Draper, D.: Model uncertainty yes, discrete model averaging maybe. Stat. Sci. 14, 405–409 (1999)
Drucker, H., Cortes, C., Jackel, L.D., LeCun, Y., Vapnik, V.: Boosting and other ensemble methods. Neural Comput. 6, 1289–1301 (1994)
Efron, B.: Estimating the error rate of a prediction rule: improvement on cross-validation. J. Am. Stat. Assoc. 78, 316–331 (1983)
Efron, B.: How biased is the apparent error rate of a prediction rule? J. Am. Stat. Assoc. 81, 461–470 (1986)
Efron, B., Tibshirani, R.: Improvements on cross-validation: the 632+ bootstrap method. J. Am. Stat. Assoc. 92, 548–560 (1997)
Efron, B.: The estimation of prediction error: covariance penalties and cross-validation. J. Am. Stat. Assoc. 99, 619–632 (2004)
Efron, B.: Estimation and accuracy after model selection. J. Am. Stat. Assoc. 109, 991–1007 (2014)
Efron, B., Hastie, T.: Computer Age Statistical Inference, vol. 5. Cambridge University Press (2016)
Ewald, K., Schneider, U.: Uniformly valid confidence sets based on the Lasso. Electron. J. Stat. 12, 1358–1387 (2018)
Fang, Y.: Asymptotic equivalence between cross-validations and Akaike information criteria in mixed-effects models. J. Data Sci. 9, 15–21 (2011)
Fletcher, D., Dillingham, P.W.: Model-averaged confidence intervals for factorial experiments. Comput. Stat. Data. An. 55, 3041–3048 (2011)
Fletcher, D., Turek, D.: Model-averaged profile likelihood intervals. J. Agr. Biol. Environ. Stat. 17, 38–51 (2011)
Fletcher, D.: Estimating overdispersion when fitting a generalized linear model to sparse data. Biometrika 99, 230–237 (2011)
Foster, D.P., George, E.I.: The risk inflation criterion for multiple regression. Ann. Stat. 22, 1947–1975 (1994)
Freund, Y.: Boosting a weak learning algorithm by majority. Inform. Comput. 121, 256–285 (1995)
Freund, Y., Schapire, R.E.: A decision-theoretic generalization of online learning and an application to boosting. J. Comput. Syst. Sci. 55, 119–139 (1997)
Fu, P., Pan, J.: A review on high-dimensional frequentist model averaging. Open. J. Sta. 8, 513–518 (2018)
Gal, Y., Ghahramani, Z.: Dropout as a Bayesian approximation: representing model uncertainty in deep learning. In: International Conference on Machine Learning, pp. 1050–1059 (2016)
Galipaud, M., Gillingham, M.A.F., David, M., Dechaume-Moncharmont, F.-X.: Ecologists overestimate the importance of predictor variables in model averaging: a plea for cautious interpretations. Methods Ecol. Evol. 5, 983–991 (2014)
Galipaud, M., Gillingham, M.A.F., DechaumeMoncharmont, F.-X.: A farewell to the sum of Akaike weights: the benefits of alternative metrics for variable importance estimations in model selection. Methods Ecol. Evol. 8, 1668–1678 (2017)
Gao, Y., Zhang, X., Wang, S., Zou, G.: Model averaging based on leave-subject-out cross-validation. J. Econ. 192, 139–151 (2016)
Gao, Y., Zhang, X., Wang, S., Chong, T.T., Zou, G. Frequentist model averaging for threshold models. Ann. I. Stat. Math. (2018). https://doi.org/10.1007/s10463-017-0642-9
Genre, V., Kenny, G., Meyler, A., Timmermann, A.: Combining expert forecasts: can anything beat the simple average? Int. J. Forecast. 29, 108–121 (2013)
George, E., Foster, D.P.: Calibration and empirical Bayes variable selection. Biometrika 87, 731–747 (2000)
Geweke, J., Amisano, G.: Optimal prediction pools. J. Econ. 164, 130–141 (2011)
Giam, X., Olden, J.D.: Quantifying variable importance in a multimodel inference framework. Methods Ecol. Evol. 7, 388–397 (2016)
Graefe, A., Kchenhoff, H., Stierle, V., Riedl, B.: Limitations of ensemble Bayesian model averaging for forecasting social science problems. Int. J. Forecast. 31, 943–951 (2015)
Greven, S., Kneib, T.: On the behaviour of marginal and conditional AIC in linear mixed models. Biometrika 97, 773–789 (2010)
Hall, S.G., Mitchell, J.: Combining density forecasts. Int. J. Forecast. 23, 1–13 (2007)
Hansen, M.H., Yu, B.: Model selection and the principle of minimum description length. J. Am. Stat. Assoc. 96, 746–774 (2001)
Hansen, B.E.: Least squares model averaging. Econometrica 75, 1175–1189 (2007)
Hansen, B.E.: Least-squares forecast averaging. J. Econ. 146, 342–350 (2008)
Hansen, B.E.: Averaging estimators for regressions with a possible structural break. Economet. Theor. 25, 1498–1514 (2009)
Hansen, P.R., Lunde, A., Nason, J.M.: The model confidence set. Econometrica 79, 453–497 (2011)
Hansen, B.E., Racine, J.S.: Jackknife model averaging. J. Econ. 167, 38–46 (2012)
Hastie, T., Tibshirani, R., Friedman, J.J.H.: The Elements of Statistical Learning, vol. 1. Springer, New York (2001)
Hauenstein, S., Wood, S.N., Dormann, C.F.: Computing AIC for black-box models using generalized degrees of freedom: a comparison with cross-validation. Commun. Stat.-Simul. C. 47, 1382–1396 (2018)
Henderson, D.J., Parmeter, C.F.: Model averaging over nonparametric estimators. In: Essays in Honor of Aman Ullah. Advances in Econometrics, vol. 36, pp. 539–560. Emerald Group Publishing Limited, UK (2016)
Hinde, J., Demtrio, C.G.B.: Overdispersion: models and estimation. Comput. Stat. Data Anal. 27, 151–170 (1998)
Hjort, N.L., Claeskens, G.: Frequentist model average estimators. J. Am. Stat. Assoc. 98, 879–945 (2003)
Hjort, N.L., Claeskens, G.: Rejoinder to the Discussion of Hjort, N.L., Claeskens, G.: Frequentist model average estimators. J. Am. Stat. Assoc. 98, 938–945 (2003)
Hjort, N.L., Claeskens, G.: Focused information criteria and model averaging for the Cox hazard regression model. J. Am. Stat. Assoc. 101, 1449–1464 (2006)
Hobbs, N.T., Hilborn, R.: Alternatives to statistical hypothesis testing in ecology: a guide to self teaching. Ecol. Appl. 16, 5–19 (2006)
Holbrook, A., Gillen, D.: Estimating prediction error for complex samples (2017). arXiv preprint: arXiv:1711.04877
Hong, C.Y.: Focussed model averaging in generalised linear models. Thesis, Doctor of Philosophy, University of Otago (2018)
Hoogerheide, L., Kleijn, R., Ravazzolo, F., Van Dijk, H.K., Verbeek, M.: Forecast accuracy and economic gains from Bayesian model averaging using time-varying weights. J. Forecast. 29, 251–269 (2010)
Hurvich, C.M., Tsai, C.-L.: Regression and time series model selection in small samples. Biometrika 76, 297–307 (1989)
Hurvich, C.M., Tsai, C.-L.: Model selection for extended quasi-likelihood models in small samples. Biometrics 51, 1077–1084 (1995)
Ishiguro, M., Sakamoto, Y.: WIC: An Estimation-free Information Criterion. Research Memorandum, Institute of Statistical Mathematics, Tokyo (1991)
Ishiguro, M., Sakamoto, Y., Kitagawa, G.: Bootstrapping log likelihood and EIC, an extension of AIC. Ann. Inst. Stat. Math. 49, 411–434 (1997)
Jacobs, R.A.: Methods for combining experts’ probability assessments. Neural Comput. 7, 867–888 (1995)
Jensen, S.M., Ritz, C.: Simultaneous inference for model averaging of derived parameters. Risk Anal. 35, 68–76 (2015)
Jiang, J., Rao, J.S., Gu, Z., Nguyen, T.: Fence methods for mixed model selection. Ann. Stat. 36, 1669–1692 (2008)
Jin, S., Ankargren, S.: Frequentist model averaging in structural equation modelling. Psychometrika (2018). https://doi.org/10.1007/s11336-018-9624-y
Johnson, W.O.: Discussion of Hjort, N.L., Claeskens, G.: Frequentist model average estimators. J. Am. Stat. Assoc. 98, 919–921 (2003)
Jullum, M., Hjort, N.L.: Parametric or nonparametric: the FIC approach. Stat. Sin. 27, 951–981 (2017)
Kabaila, P., Leeb, H.: On the large-sample minimal coverage probability of confidence intervals after model selection. J. Am. Stat. Assoc. 101, 619–629 (2006)
Kabaila, P., Welsh, A.H., Abeysekera, W.: Model-averaged confidence intervals. Scand. J. Stat. 43, 35–48 (2016)
Kabaila, P., Welsh, A.H., Mainzer, R.: The performance of model averaged tail area confidence intervals Commun. Stat-Theor. M. 46, 10718–10732 (2016)
Kabaila, P., Wijethunga, C.: Confidence intervals centered on bootstrap smoothed estimators (2016). arXiv preprint: arXiv:1610.09802
Kabaila, P.: On the minimum coverage probability of model averaged tail area confidence intervals. Can. J. Stat. 46, 279–297 (2018)
Kapetanios, G., Mitchell, J., Price, S., Fawcett, N.: Generalised density forecast combinations. J. Econ. 188, 150–165 (2015)
LeBlanc, M., Tibshirani, R.: Combining estimates in regression and classification. J. Am. Stat. Assoc. 91, 1641–1650 (1996)
Lee, H., Jogesh Babu, G., Rao, C.R.R.: A jackknife type approach to statistical model selection. J. Stat. Plan. Inference 142, 301–311 (2012)
Leeb, H., Pötscher, B.M.: Model selection and inference: facts and fiction. Econ. Theory 21, 21–59 (2005)
Leeb, H., Pötscher, B.M.: Can one estimate the conditional distribution of post-model-selection estimators? Ann. Stat. 34, 2554–2591 (2006)
Lemke, C., Budka, M., Gabrys, B.: Metalearning: a survey of trends and technologies. Artif. Intell. Rev. 44, 117–130 (2015)
Lenkoski, A., Eicher, T.S., Raftery, A.E.: Two-stage Bayesian model averaging in endogenous variable models. Econ. Rev. 33, 122–151 (2014)
Leung, G., Barron, A.R.: Information theory and mixing least-squares regressions. IEEE Trans. Inf. Theory 52, 3396–3410 (2006)
Li, C., Li, Q., Racine, J.S., Zhang, D.: Optimal model averaging of varying-coefficient models. Stat. Sin. (2018). https://doi.org/10.5705/ss.202017.0034
Li, J., Xia, X., Wong, W.K., Nott, D.: Varying-coefficient semiparametric model averaging prediction. Biometrics (2018). https://doi.org/10.1111/biom.12904
Liang, H., Wu, H., Zou, G.: A note on conditional AIC for linear mixed-effects models. Biometrika 95, 773–778 (2008)
Liang, H., Zou, G., Wan. A.T.K., Zhang, X.: Optimal weight choice for frequentist model average estimators: J. Am. Stat. Assoc. 106, 1053–1066 (2011)
Lieb, L., Smeekes, S.: Inference for impulse responses under model uncertainty (2017). arXiv preprint: arXiv:1709.09583
Lin, B., Wang, Q., Zhang, J., Pang, Z.: Stable prediction in high-dimensional linear models. Stat. Comput. 27, 1401–1412 (2017)
Link, W., Barker, R.: Model weights and the foundations of multimodel inference. Ecology 87, 2626–2635 (2006)
Liu, Q., Okui, R.: Heteroscedasticity-robust \(\text{ C }_{\text{ p }}\) model averaging. Econ. J. 16, 463–472 (2013)
Liu, S., Yang, Y.: Combining models in longitudinal data analysis. Ann. Inst. Stat. Math. 64, 233–254 (2012)
Liu, S., Yang, Y.: Mixing partially linear regression models. Sankhyā 75, 74–95 (2013)
Liu, C.A.: Distribution theory of the least squares averaging estimator. J. Econ. 186, 142–159 (2015)
Liu, Q., Okui, R., Yoshimura, A.: Generalized least squares model averaging. Econ. Rev. 35, 1692–1752 (2016)
Longford, N.T.: An alternative to model selection in ordinary regression. Stat. Comput. 13, 67–80 (2003)
Longford, N.T.: An alternative analysis of variance. SORT Stat. Oper. Res. T. 32, 77–92 (2008)
Lu, X., Su, L.: Jackknife model averaging for quantile regressions. J. Econ. 188, 40–58 (2015)
Lumley, T., Scott, A.: AIC and BIC for modeling with complex survey data. J. Surv. Stat. Methodol. 3, 1–18 (2015)
Lv, J., Liu, J.S.: Model selection principles in misspecified models. J. R. Stat. Soc. Ser. B (Stat. Methodol.) 76, 141–167 (2014)
Magnus, J.R., Wan, A.T.K., Zhang, X: Weighted average least squares estimation with nonspherical disturbances and an application to the Hong Kong housing market. Comput. Stat. Data Anal. 55, 1331–1341 (2011)
Magnus, J.R., De Luca, G.: Weighted-average least squares (WALS): a survey. J. Econ. Surv. 30, 117–148 (2016)
Mallows, C.L.: Some comments on Cp. Technometrics 42, 87–94 (2000)
Martins, L.F., Gabriel, V.J.: Linear instrumental variables model averaging estimation. Comput. Stat. Data. Anal. 71, 709–724 (2014)
McQuarrie, A., Shumway, R., Tsai, C.-L.: The model selection criterion AICu. Stat. Probabil. Lett. 34, 285–292 (1997)
McQuarrie, A.D.R., Tsai, C.-L.: Regression and Time Series Model Selection. World Scientific, Singapore (1998)
Mead, R.: The Design of Experiments: Statistical Principles for Practical Applications. Cambridge University Press, Cambridge (1988)
Mitra, P., Lian, H., Mitra, R., Liang, H., Xie, M.: A general framework for frequentist model averaging (2018). arXiv preprint: arXiv:1802.03511
Moody, J.E.: The effective number of parameters: an analysis of generalization and regularization in nonlinear learning systems. In: Moody, J.E., Hanson, S.J., Lippmann, R.P. (eds.) Advances in Neural Information Processing Systems, vol. 4, pp. 847–854. Morgan Kaufmann, San Mateo, California (1992)
Müller, S., Scealy, J.L., Welsh, A.H.: Model selection in linear mixed models. Stat. Sci. 28, 135–167 (2013)
Murata, N., Yoshizawa, S., Amari, S.: Network information criterion-determining the number of hidden units for artificial neural network models. IEEE Trans. Neural Netw. 5, 865–872 (1994)
Murray, K., Conner, M.M.: Methods to quantify variable importance: implications for the analysis of noisy ecological data. Ecology 90, 348–355 (2009)
Naftaly, U., Intrator, N., Horn, D.: Optimal ensemble averaging of neural networks. Network-Comp. Neural 8, 283–296 (1997)
Nakagawa, S., Freckleton, R.P.: Model averaging, missing data and multiple imputation: a case study for behavioural ecology. Behav. Ecol. Sociobiol. 65, 103–116 (2011)
Neath, A.A., Cavanaugh, J.E., Weyhaupt, A.G.: Model evaluation, discrepancy function estimation, and social choice theory. Comput. Stat. 29, 1–19 (2014)
Owen, A.B.: Small sample central confidence intervals for the mean. Technical Report 302, Department of Statistics, Stanford University (1988)
Polley, E.C., van der Laan, M.J.: Super learner in prediction. UC Berkeley Division of Biostatistics Working Paper Series. Working Paper 266 (2010). http://biostats.bepress.com/ucbbiostat/paper266
Poeter, E.P., Hill, M.C.: MMA, a computer code for multi-model analysis. U.S. Geological Survey Techniques and Methods TM6-E3. Reston, Virginia (2007)
Pötscher, B.M.: The distribution of model averaging estimators and an impossibility result regarding its estimation. Inst. Math. S. 52, 113–129 (2006)
Quenouille, M.H.: Notes on bias in estimation. Biometrika 43, 353–360 (1956)
R Core Team: R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria (2017). https://www.R-project.org/
Raftery, A.E., Zheng, Y.: Discussion of Hjort, N.L., Claeskens, G.: Frequentist model average estimators. J. Am. Stat. Assoc. 98, 931–938 (2003)
Rao, J.S., Tibshirani, R.: The out-of-bootstrap method for model averaging and selection. University of Toronto (1997)
Richards, S.A.: Testing ecological theory using the information-theoretic approach: examples and cautionary results. Ecology 86, 2805–2814 (2005)
Ripley, B.D.: Selecting amongst large classes of models. In: Adams, N., Crowder, M., Hand, D.J., Stephens, D. (eds.) Methods and Models in Statistics: in Honor of Professor John Nelder, FRS, pp. 155–170. Imperial College Press, London (2004)
Rubin, D.B.: Inference and missing data. Biometrika 63, 581–592 (1976)
Saefken, B., Kneib, T., van Waveren, C.-S., Greven, S.: A unifying approach to the estimation of the conditional Akaike information in generalized linear mixed models. Electron. J. Stat. 8, 201–225 (2014)
Sapp, S., van der Laan, M.J., Canny, J.: Subsemble: an ensemble method for combining subset-specific algorithm fits. J. Appl. Stat. 41, 1247–1259 (2014)
Schapire, R.E.: The strength of weak learnability. Mach. Learn. 5, 197–227 (1990)
Schomaker, M.: Shrinkage averaging estimation. Stat. Pap. 53, 1015–1034 (2012)
Schomaker, M., Wan, A.T.K., Heumannm, C.: Frequentist model averaging with missing observations. Comput. Stat. Data. Anal. 54, 3336–3347 (2010)
Schomaker, M., Heumannm, C.: Model selection and model averaging after multiple imputation. Comput. Stat. Data. Anal. 71, 758–770 (2014)
Schomaker, M., Heumann, C.: When and when not to use optimal model averaging (2018). arXiv preprint: arXiv:1802.04589
Shan, K., Yang, Y.: Combining regression quantile estimators. Stat. Sin. 19, 1171–1191 (2009)
Shang, J., Cavanaugh, J.E.: Bootstrap variants of the Akaike information criterion for mixed model selection. Comput. Stat. Data. Anal. 52, 2004–2021 (2008)
Shen, X., Huang, H.-C., Ye, J.: Adaptive model selection and assessment for exponential family distributions. Technometrics 46, 306–317 (2004)
Shen, X., Huang, Huang.-C.: Optimal model assessment, selection, and combination. J. Am. Stat. Assoc. 101, 554–568 (2006)
Shibata, R.: Bootstrap estimate of Kullback-Leibler information for model selection. Stat. Sin. 7, 375–394 (1997)
Smith, A.C., Koper, N., Francis, C.M., Fahrig, L.: Confronting collinearity: comparing methods for disentangling the effects of habitat loss and fragmentation. Landscape Ecol. 24, 1271–1285 (2009)
Smyth, P., Wolpert, D.: Linearly combining density estimators via stacking. Mach. Learn. 36, 59–83 (1999)
Stone, M.: Cross-validatory choice and assessment of statistical predictions. J. R. Stat. Soc. Ser. B. (Methodol.) 36, 111–147 (1974)
Stone, M.: An asymptotic equivalence of choice of model by cross-validation and Akaike’s criterion. J. R. Stat. Soc. Ser. B. (Methodol.) 39, 44–47 (1977)
Sugiura, N.: Further analysts of the data by Akaike’s information criterion and the finite corrections: further analysts of the data by Akaike’s. Commun. Stat. Theory 7, 13–26 (1978)
Takeuchi, K.: Distribution of informational statistics and a criterion of model fitting. Suri-Kagaku (Math. Sci.) 153, 12–18 (1976)
Timmermann, A.: Forecast combinations. In: Elliott, G., Granger, C.W.J., Timmermann, A. (eds.) Handbook of Economic Forecasting, pp. 135–196. Elsevier, Amsterdam (2006)
Ting, K.M., Witten, I.H.: Issues in stacked generalization. J. Artif. Intell. Res. 10, 271–289 (1999)
Turek, D., Fletcher, D.: Model-averaged Wald confidence intervals. Comput. Stat. Data. Anal. 56, 2809–2815 (2012)
Turek, D.: Comparison of the frequentist MATA confidence interval with Bayesian model-averaged confidence intervals. J. Probab. Stat. (2015). https://doi.org/10.1155/2015/420483
Ullah, A., Wang, H.: Parametric and nonparametric frequentist model selection and model averaging. Econ. J. 1, 157–179 (2013)
Vaida, F., Blanchard, S.: Conditional Akaike information for mixed-effects models. Biometrika 92, 351–370 (2005)
van der Laan, M.J., Dudoit, S., Keles, S.: Asymptotic optimality of likelihood-based cross-validation. Stat. Appl. Genet. Mol. 3, Article 4 (2004)
van der Laan, M.J., Polley, E.C., Hubbard, A.E.: Super learner. Stat. Appl. Genet. Mol. Biol. 6, 1–23 (2007)
Wagenmakers, E.-J., Farrell, S., Ratcliff, R.: Letter to the editors. Biometrics 60, 281–283 (2004)
Wager, S., Hastie, T., Efron, B.: Confidence intervals for random forests: the jackknife and the infinitesimal jackknife. J. Mach. Learn. Res. 15, 1625–1651 (2014)
Wallis, K.F.: Combining density and interval forecasts: a modest proposal. Oxford B. Econ. Stat. 67, 983–994 (2005)
Wan, A.T.K., Zhang, X., Zou, G.: Least squares model averaging by Mallows criterion. J. Econ. 156, 277–283 (2010)
Wan, A.T.K., Zhang, X., Wang, S.: Frequentist model averaging for multinomial and ordered logit models. In. J. Forecast. 30, 118–128 (2014)
Wang, H., Zou, G., Wan, A.T.K.: Model averaging for varying-coefficient partially linear measurement error models. Electron. J. Stat. 6, 1017–1039 (2012)
Wang, H., Zhou, S.Z.F.: Interval estimation by frequentist model averaging. Commun. Stat. Theory 42, 4342–4356 (2013)
Wang, H.Y., Chen, X., Flournoy, N.: The focused information criterion for varying-coefficient partially linear measurement error models. Stat. Pap. 1–15. Springer, Heidelberg (2014)
Wang, H., Li, Y., Sun, J.: Focused and model average estimation for regression analysis of panel count data. Scand. J. Stat. 42, 732–745 (2015)
Wedderburn, R.W.M.: Quasi-likelihood functions, generalized linear models, and the Gauss-Newton method. Biometrika 61, 439–447 (1974)
White, H.: Maximum likelihood estimation of misspecified models. Econometica 50, 1–25 (1982)
Wolpert, D.H.: Stacked generalization. Neural Netw. 5, 241–259 (1992)
Wood, S.N.: Core Statistics. Cambridge University Press, Cambridge (2015)
Xie, T.: Prediction model averaging estimator. Econ. Lett. 131, 5–8 (2015)
Xu, R., Gamst, A., Donohue, M., Vaida, F., Harrington, D.P.: Using profile likelihood for semiparametric model selection with application to proportional hazards mixed models. Harvard University Biostatistics Working Paper Series, Paper 43 (2006). http://biostats.bepress.com/harvardbiostat/paper43/
Xu, G., Wang, S., Huang, J.Z.: Focused information criterion and model averaging based on weighted composite quantile regression. Scand. J. Stat. 41, 365–381 (2014)
Xu, R., Mehrotra, D.V., Shaw, P.A.: Incorporating baseline measurements into the analysis of crossover trials with timetoevent endpoints. Stat. Med. (2018). https://doi.org/10.1002/sim.7834
Yang, Y.: Adaptive regression by mixing. J. Am. Stat. Assoc. 96, 574–588 (2001)
Yang, Y.: Regression with multiple candidate models: selecting or mixing? Stat. Sin. 13, 783–809 (2003)
Yang, Y.: Can the strengths of AIC and BIC be shared? A conflict between model indentification and regression estimation. Biometrika 92, 937–950 (2005)
Ye, J.: On measuring and correcting the effects of data mining and model selection. J. Am. Stat. Assoc. 93, 120–131 (1998)
Yu, D., Yau, K.K.W.: Conditional Akaike information criterion for generalized linear mixed models. Comput. Stat. Data. Anal. 56, 629–644 (2012)
Yu, Y., Thurston, S.W., Hauser, R., Liang, H.: Model averaging procedure for partially linear single-index models. J. Stat. Plan. Infer. 143, 2160–2170 (2013)
Yu, W., Xu, W., Zhu, L.: Transformation-based model averaged tail area inference. Comput. Stat. 29, 1713–1726 (2014)
Yu, D., Zhang, X., Yau, K.K.W.: Asymptotic properties and information criteria for misspecified generalized linear mixed models. J. R. Stat. Soc. Ser. B (Methodol.) (2018). https://doi.org/10.1111/rssb.12270
Yuan, Z., Yang, Y.: Combining linear regression models. J. Am. Stat. Assoc. 100, 1202–1214 (2005)
Yuan, Z., Ghosh, D.: Combining multiple biomarker models in logistic regression. Biometrics 64, 431–439 (2008)
Zeng, J.: Model-Averaged Confidence Intervals. (Thesis, Doctor of Philosophy). University of Otago (2013)
Zeng, J., Cheng, W., Hu, G., Ronga, Y.: Model averaging procedure for varying-coefficient partially linear models with missing responses. J. Korean Stat. Soc. 47, 379–394 (2018)
Zhang, X., Liang, H.: Focused information criterion and model averaging for generalized additive partial linear models. Ann. Stat. 39, 174–200 (2011)
Zhang, C., Ma, Y.: (eds.) Ensemble Machine Learning: Methods and Applications. Springer, New York (2012)
Zhang, X., Wan, A.T.K., Zhou, S.Z.: Focused information criteria, model selection, and model averaging in a Tobit model with a nonzero threshold. J. Bus. Econ. Stat. 30, 132–142 (2012)
Zhang, X., Wan, A.T.K., Zou, G.: Model averaging by jackknife criterion in models with dependent data. J. Econ. 174, 82–94 (2013)
Zhang, X., Zou, G., Carroll, R.J.: Model averaging based on Kullback-Leibler distance. Stat. Sin. 25, 1583–1598 (2015)
Zhang, X.: Consistency of model averaging estimators. Econ. Lett. 130, 120–123 (2015)
Zhang, Y., Yang, Y.: Cross-validation for selecting a model selection procedure. J. Econ. 187, 95–112 (2015)
Zhang, X., Yu, D., Zou, G., Liang, H.: Optimal model averaging estimation for generalized linear models and generalized linear mixed-effects models. J. Am. Stat. Assoc. 111, 1775–1790 (2016)
Zhang, Q., Duan, X., Ma, S.: Focused information criterion and model averaging with generalized rank regression. Stat. Probabil. Lett. 122, 11–19 (2017)
Zhao, N., Zhao, Z., Liao, S.: Probabilistic model combination for support vector machine using positive-definite kernel-based regularization path. In: Wang, Y., Li, T. (eds.) Foundations of Intelligent Systems. Advances in Intelligent and Soft Computing, vol. 122, pp. 201–206. Springer, Heidelberg (2011)
Zhao, S., Zhang, X., Gao, Y.: Model averaging with averaging covariance matrix. Econ. Lett. 145, 214–217 (2016)
Zhao, S., Ullah, A., Zhang, X.: A class of model averaging estimators. Econ. Lett. 162, 101–106 (2018)
Zou, G., Wan, A.T.K., Wu, X., Chen, T.: Estimation of regression coefficients of interest when other regression coefficients are of no interest: the case of non-normal errors. Stat. Probabil. Lett. 77, 803–810 (2007)
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
Copyright information
© 2018 The Author(s), under exclusive licence to Springer-Verlag GmbH, DE, part of Springer Nature
About this chapter
Cite this chapter
Fletcher, D. (2018). Frequentist Model Averaging. In: Model Averaging. SpringerBriefs in Statistics. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-662-58541-2_3
Download citation
DOI: https://doi.org/10.1007/978-3-662-58541-2_3
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-662-58540-5
Online ISBN: 978-3-662-58541-2
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)