Abstract
The prequential approach to statistics leads naturally to model list selection because the sequential reformulation of the problem is a guided search over model lists drawn from a model space. That is, continually updating the action space of a decision problem to achieve optimal prediction forces the collection of models under consideration to grow neither too fast nor too slow to avoid excess variance and excess bias, respectively. At the same time, the goal of good predictive performance forces the search over good predictors formed from a model list to close in on the data generator. Taken together, prequential model list re-selection favors model lists which provide an effective approximation to the data generator but do so by making the approximation match the unknown function on important regions as determined by empirical bias and variance.
Similar content being viewed by others
References
Alonso A, Pena D, Romo J (2006) Introducing model uncertainty by moving blocks bootstrap. Stat Papers 47: 167–179
Berger J, Barbieri M (2004) Optimal predictive model selection. Ann Stat 32: 870–897
Chen SS (1995) Basis pursuit. Ph.D. thesis, Department of Statistics, Stanford University. http://www-stat.stanford.edu/schen
Chen SS, Donoho DL, Saunders MA (1998) Atomic decomposition by basis pursuit. SIAM J Sci Comput 20: 33–61
Chen SS, Donoho DL, Saunders MA (2001) Atomic decomposition by basis pursuit. SIAM Rev 43: 129–159
Clarke J, Clarke B (2009) Prequential analysis of complex data. Stat Anal Data Min (to appear)
Czado C, Raftery A (2006) Choosing the link function and accounting for link uncertainty in generalized linear models using Bayes factors. Stat Papers 47: 419–442
Daubechies I (1988) Time-frequency localization operators: a geometric phase space approach. IEEE Trans Inf Theory 34: 605–612
Dawid AP (1984) Present position and potential developments: some personal views. Statsitical theory. The prequential approach. J R Stat Soc B 147: 278–292
Dawid AP, Vovk V (1999) Prequential probability: principles and properties. Bernoulli 5: 125–162
Draper D (1995) Assessment and propagation of model uncertainty. J R Stat Soc B 57: 45–97
George EI (2000) The variable selection problem. J Am Stat Assoc 95: 1304–1308
George EI, McCulloch RE (1993) Variable selection via Gibbs sampling. J Am Stat Assoc 88: 881–889
Gustafson P, Clarke B (2004) Decomposing posterior variance. J Stat Plan Inference 119: 311–327
Hjort NL, Claeskens G (2003) Frequentist model average estimators. J Am Stat Assoc 98: 879–899
Hoeting JA, Madigan D, Raftery AE, Volinsky CT (1999) Bayesian model averaging: a tutorial. Stat Sci 14: 382–401
Kass R, Raftery A (1995) Bayes factors. J Am Stat Assoc 90: 773–795
Leamer E (1978) Specification searches, ad hoc inference with nonexperimental data. Wiley, New York
Minka T (2000) Bayes model averaging is not model combination. Technical report. http://citeseerx.ist.psu.edu
Raftery AE, Madigan D, Hoeting JA (1997) Bayesian model averaging for linear regression models. J Am Stat Assoc 92: 179–191
Toutenberg H, Toutenberg H (2002) Prediction of response values in linear regression models from replicated experiments. Stat Papers 43: 423–433
Wolpert D (1992) Stacked generalization. Neural Netw 5: 241–259
Wong A (1995) An averaging approach to prediction. Stat Papers 36: 253–264
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Fokoue, E., Clarke, B. Bias-variance trade-off for prequential model list selection. Stat Papers 52, 813–833 (2011). https://doi.org/10.1007/s00362-009-0289-6
Received:
Revised:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00362-009-0289-6