Advertisement

Sequential Super Learning

  • Sherri Rose
  • Mark J. van der Laan
Chapter
Part of the Springer Series in Statistics book series (SSS)

Abstract

Suppose a doctor is interested in predicting the individual outcomes for a group of patients under two particular drug regimens at two time points in the future. She is therefore asking, what would happen to each of these patient’s future outcomes at these time points if I were to enforce drug regimen 1 or drug regimen 2? Which treatment will be better for the patients’ efficacy outcomes? Which treatment will be better for the patients’ safety outcomes? Prediction problems can be longitudinal in nature, generally, and we frequently wish to understand what the mean outcome of patients with certain characteristics would be months or years in the future. Often, this is under the setting where we would hypothetically assign a particular treatment “rule” to the patients.

References

  1. J.C. Brooks, Super learner and targeted maximum likelihood estimation for longitudinal data structures with applications to atrial fibrillation. PhD thesis, University of California, Berkeley (2012)Google Scholar
  2. I. Díaz, A. Hubbard, A. Decker, M. Cohen, Variable importance and prediction methods for longitudinal problems with missing variables. PLoS One 10(3), e0120031 (2015)Google Scholar
  3. S. Dudoit, M.J. van der Laan, Asymptotics of cross-validated risk estimation in estimator selection and performance assessment. Stat. Methodol. 2(2), 131–154 (2005)MathSciNetCrossRefzbMATHGoogle Scholar
  4. S. Keleş, M.J. van der Laan, S. Dudoit, Asymptotically optimal model selection method for regression on censored outcomes. Technical Report, Division of Biostatistics, University of California, Berkeley (2002)zbMATHGoogle Scholar
  5. M.L. Petersen, E. LeDell, J. Schwab, V. Sarovar, R. Gross, N. Reynolds, J.E. Haberer, K. Goggin, C. Golin, J. Arnsten et al., Super learner analysis of electronic adherence data improves viral prediction and may provide strategies for selective HIV RNA monitoring. J. Acquir. Immune Defic. Syndr. 69(1), 109 (2015)Google Scholar
  6. R. Pirracchio, M.L. Petersen, M. Carone, M.R. Rigon, S. Chevret, M.J. van der Laan, Mortality prediction in intensive care units with the super ICU learner algorithm (SICULA): a population-based study. Lancet Respir. Med. 3(1), 42–52 (2015)CrossRefGoogle Scholar
  7. E.C. Polley, M.J. van der Laan, Super learner in prediction. Technical Report 266, Division of Biostatistics, University of California, Berkeley (2010)Google Scholar
  8. E.C Polley, S. Rose, M.J. van der Laan, Super-learning, in Targeted Learning: Causal Inference for Observational and Experimental Data, ed. by M.J. van der Laan, S. Rose (Springer, Berlin, Heidelberg, New York, 2011)Google Scholar
  9. J.M. Robins, Robust estimation in sequentially ignorable missing data and causal inference models, in Proceedings of the American Statistical Association (2000)Google Scholar
  10. J.M. Robins, Y. Ritov, Toward a curse of dimensionality appropriate (coda) asymptotic theory for semi-parametric models. Stat. Med. 16, 285–319 (1997)CrossRefGoogle Scholar
  11. S. Rose, Mortality risk score prediction in an elderly population using machine learning. Am. J. Epidemiol. 177(5), 443–452 (2013)CrossRefGoogle Scholar
  12. S. Rose, A machine learning framework for plan payment risk adjustment. Health Serv. Res. 51(6), 2358–2374 (2016)CrossRefGoogle Scholar
  13. M.J. van der Laan, S. Dudoit, Unified cross-validation methodology for selection among estimators and a general cross-validated adaptive epsilon-net estimator: finite sample oracle inequalities and examples. Technical Report, Division of Biostatistics, University of California, Berkeley (2003)Google Scholar
  14. M.J. van der Laan, E.C. Polley, A.E. Hubbard, Super learner. Stat. Appl. Genet. Mol. 6(1), Article 25 (2007)Google Scholar
  15. M.J. van der Laan, S. Dudoit, S. Keleş, Asymptotic optimality of likelihood-based cross-validation. Stat. Appl. Genet. Mol. 3(1), Article 4 (2004)Google Scholar
  16. M.J. van der Laan, S. Dudoit, A.W. van der Vaart. The cross-validated adaptive epsilon-net estimator. Stat. Decis. 24(3), 373–395 (2006)MathSciNetzbMATHGoogle Scholar
  17. A.W. van der Vaart, S. Dudoit, M.J. van der Laan, Oracle inequalities for multi-fold cross-validation. Stat. Decis. 24(3), 351–371 (2006)MathSciNetzbMATHGoogle Scholar

Copyright information

© Springer International Publishing AG 2018

Authors and Affiliations

  1. 1.Department of Health Care PolicyHarvard Medical SchoolBostonUSA
  2. 2.Division of Biostatistics and Department of StatisticsUniversity of California, BerkeleyBerkeleyUSA

Personalised recommendations