Abstract
In this chapter, we illustrated the problem of non-regularity that arises in the context of inference about the optimal current treatment rule, when the optimal treatments at subsequent stages are non-unique for at least some strictly positive proportion of subjects in the population. We discuss and illustrate the phenomenon using Q-learning and G-estimation, and propose a number of strategies to mitigate the non-regularity including thresholding and penalization of the estimators as well as non-standard implementations of the bootstrap.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
This is unpublished work, but the first author was pointed to this direction by Dr. Susan Murphy (personal communication).
- 2.
If this unlikely event does occur, one should examine the observed values of \(\hat{p}\). If the values of \(\hat{p}\) are concentrated close to zero, ν may be increased; if not, the maximal value in the grid should be increased.
References
Andrews, D. W. K. (2000). Inconsistency of the bootstrap when a parameter is on the boundary of the parameter space. Econometrica, 68, 399–405.
Arjas, E., & Saarela, O. (2010). Optimal dynamic regimes: Presenting a case for predictive inference. The International Journal of Biostatistics, 6.
Arroll, B., MacGillivray, S., Ogston, S., Reid, I., Sullivan, F., Williams, B., & Crombie, I. (2005). Efficacy and tolerability of tricyclic antidepressants and ssris compared with placebo for treatment of depression in primary care: A meta-analysis. Annals of Family Medicine, 3, 449–456.
Arroll, B., Elley, C. R., Fishman, T., Goodyear-Smith, F. A., Kenealy, T., Blashki, G., Kerse, N., & MacGillivray, S. (2009). Antidepressants versus placebo for depression in primary care. Cochrane Database of Systematic Reviews, 3, CD007954.
Berger, R. L. (1996). More powerful tests from confidence interval p values. American Statistician, 50, 314–318.
Berger, R. L., & Boos, D. D. (1994). P values maximized over a confidence set for the nuisance parameter. Journal of the American Statistical Association, 89, 1012–1016.
Bickel, P. J., & Sakov, A. (2008). On the choice of m in the m out of n bootstrap and confidence bounds for extrema. Statistica Sinica, 18, 967–985.
Bickel, P. J., Klaassen, C. A. J., Ritov, Y., & Wellner, J. A. (1993). Efficient and adaptive estimation for semiparametric models. Baltimore: Johns Hopkins University Press.
Bickel, P. J., Gotze, F., & Zwet, W. V. (1997). Resampling fewer than n observations: Gains, losses and remedies for losses. Statistica Sinica, 7, 1–31.
Breiman, L. (1996). Bagging predictors. Machine Learning, 24, 123–140.
Brotman, R. M., Klebanoff, M. A., Nansel, T. R., Andrews, W. W., Schwebke, J. R., Zhang, J., Yu, K. F., Zenilman, J. M., & Scharfstein, D. O. (2008). A longitudinal study of vaginal douching and bacterial vaginosis – A marginal structural modeling analysis. American Journal of Epidemiology, 168, 188–196.
Cain, L. E., Robins, J. M., Lanoy, E., Logan, R., Costagliola, D., & Hernán, M. A. (2010). When to start treatment? A systematic approach to the comparison of dynamic regimes using observational data. The International Journal of Biostatistics, 6.
Chakraborty, B., Laber, E. B., & Zhao, Y. (2013). Inference for optimal dynamic treatment regimes using an adaptive m-out-of-n bootstrap scheme. Biometrics, (in press).
Chapman, G. B., & Sonnenberg, F. B. (2000). Decision making in health care: Theory, psychology, and applications. Cambridge, UK: Cambridge University Press.
Chow, S. C., & Chang, M. (2008). Adaptive design methods in clinical trials – A review. Orphanet Journal of Rare Diseases, 3.
Dawid, A. P., & Didelez, V. (2010). Identifying the consequences of dynamic treatment strategies: A decision-theoretic overview. Statistics Surveys, 4, 184–231.
Dragalin, V. (2006). Adaptive designs: Terminology and classification. Drug Information Journal, 40, 425–435.
Efron, B., & Tibshirani, R. (1993). An introduction to the bootstrap (Vol. 57). London: Chapman & Hall/CRC.
Ernst, D., Geurts, P., & Wehenkel, L. (2005). Tree-based batch mode reinforcement learning. Journal of Machine Learning Research, 6, 503–556.
Fava, M., Rush, A. J., Trivedi, M. H., Nierenberg, A. A., Thase, M. E., Sackeim, H. A., Quitkin, F. M., Wisniewski, S., Lavori, P. W., Rosenbaum, J. F., & Kupfer, D. J. (2003). Background and rationale for the Sequenced Treatment Alternatives to Relieve Depression (STAR*D) study. Psychiatric Clinics of North America, 26, 457–494.
Figueiredo, M., & Nowak, R. (2001). Wavelet-based image estimation: An empirical Bayes approach using Jeffreys’ noninformative prior. IEEE Transactions on Image Processing, 10, 1322–1331.
Freedman, B. (1987). Equipoise and the ethics of clinical research. The New England Journal of Medicine, 317, 141–145.
French, S. (1986). Decision theory: An introduction to the mathematics of rationality. Chichester: Ellis Horwood.
Geurts, P., Ernst, D., & Wehenkel, L. (2006). Extremely randomized trees. Machine Learning, 11, 3–42.
Hastie, T., Tibshirani, R., & Friedman, J. (2009). The elements of statistical learning: Data mining, inference, and prediction (2nd ed.). New York: Springer.
Hernán, M. A., & Robins, J. M. (2013). Causal inference. Chapman & Hall/CRC (in revision).
Laber, E. B., Qian, M., Lizotte, D., & Murphy, S. A. (2011). Statistical inference in dynamic treatment regimes. arXiv:1006.5831v2 [stat.ME].
Lavori, P. W., & Dawson, R. (2000). A design for testing clinical strategies: Biased adaptive within-subject randomization. Journal of the Royal Statistical Society, Series A, 163, 29–38.
Lee, M.-J., & Huang, F. (2012). Finding dynamic treatment effects under anticipation: The effects of spanking on behaviour. Journal of the Royal Statistical Society, Series A, 175, 535–567.
Lei, H., Nahum-Shani, I., Lynch, K., Oslin, D., & Murphy, S. A. (2012). A SMART design for building individualized treatment sequences. The Annual Review of Psychology, 8, 21–48.
Lusted, L. B. (1968). Introduction to medical decision making. Springfield: Thomas.
McEvoy, J. P., Lieberman, J. A., Stroup, T. S., Davis, S., Meltzer, H. Y., Rosenheck, R. A., Swartz, M. S., Perkins, D. O., Keefe, R. S. E., Davis, C. E., Severe, J., & Hsiao, J. K. (2006). Effectiveness of clozapine versus olanzapine, quetiapine and risperidone in patients with chronic schizophrenia who did not respond to prior atypical antipsychotic treatment. American Journal of Psychiatry, 163, 600–610.
Moodie, E. E. M. (2009b). Risk factor adjustment in marginal structural model estimation of optimal treatment regimes. Biometrical Journal, 51, 774–788.
Moodie, E. E. M., Richardson, T. S., & Stephens, D. A. (2007). Demystifying optimal dynamic treatment regimes. Biometrics, 63, 447–455.
Nelson, J. C. (1997). Safety and tolerability of the new antidepressants. Journal of Clinical Psychiatry, 58(Suppl. 6), 26–31.
Neugebauer, R., & Van der Laan, M. J. (2005). Why prefer double robust estimators in causal inference? Journal of Statistical Planning and Inference, 129, 405–426.
Neyman, J. (1923). On the application of probability theory to agricultural experiments. Essay in principles. Section 9 (translation published in 1990). Statistical Science, 5, 472–480.
Orellana, L., Rotnitzky, A., & Robins, J. M. (2010a). Dynamic regime marginal structural mean models for estimation of optimal dynamic treatment regimes, part I: Main content. The International Journal of Biostatistics, 6.
Pliskin, J. S., Shepard, D., & Weinstein, M. C. (1980). Utility functions for life years and health status: Theory, assessment, and application. Operations Research, 28, 206–224.
Pötscher, B. M. (2007). Confidence sets based on sparse estimators are necessarily large. Arxiv preprint arXiv:0711.1036.
Pötscher, B. M., & Schneider, U. (2008). Confidence sets based on penalized maximum likelihood estimators. Mpra paper, University Library of Munich, Germany.
Qian, M., & Murphy, S. A. (2011). Performance guarantees for individualized treatment rules. Annals of Statistics, 39, 1180–1210.
Robins, J. M., & Hernán, M. A. (2009). Estimation of the causal effects of time-varying exposures. In G. Fitzmaurice, M. Davidian, G. Verbeke, & G. Molenberghs (Eds.), Longitudinal data analysis. Boca Raton: Chapman & Hall/CRC.
Shao, J., & Sitter, R. R. (1996). Bootstrap for imputed survey data. Journal of the American Statistical Association, 91, 1278–1288.
Sox, H. C., Blatt, M. A., Higgins, M. C., & Marton, K. I. (1988). Medical decision making. Boston: Butterworth-Heinemann.
Thall, P. F., Sung, H. G., & Estey, E. H. (2002). Selecting therapeutic strategies based on efficacy and death in multicourse clinical trials. Journal of the American Statistical Association, 97, 29–39.
Thall, P. F., Wooten, L. H., Logothetis, C. J., Millikan, R. E., & Tannir, N. M. (2007a). Bayesian and frequentist two-stage treatment strategies based on sequential failure times subject to interval censoring. Statistics in Medicine, 26, 4687–4702.
Thall, P. F., Logothetis, C., Pagliaro, L. C., Wen, S., Brown, M. A., Williams, D., & Millikan, R. E. (2007b). Adaptive therapy for androgen-independent prostate cancer: A randomized selection trial of four regimens. Journal of the National Cancer Institute, 99, 1613–1622.
Van der Laan, M. J., & Petersen, M. L. (2007a). Causal effect models for realistic individualized treatment and intention to treat rules. The International Journal of Biostatistics, 3.
Van der Laan, M. J., & Rubin, D. (2006). Targeted maximum likelihood learning. The International Journal of Biostatistics, 2.
Vansteelandt, S., & Goetghebeur, E. (2003). Causal inference with generalized structural mean models. Journal of the Royal Statistical Society, Series B, 65, 817–835.
Wahed, A. S., & Tsiatis, A. A. (2006). Semiparametric efficient estimation of survival distributions in two-stage randomisation designs in clinical trials with censored data. Biometrika, 93, 163–177.
Wald, A. (1949). Statistical decision functions. New York: Wiley.
Zhang, T. (2004). Statistical behavior and consistency of classification methods based on convex risk minimization. Annals of Statistics, 32, 56–85.
Zou, H. (2006). The adaptive lasso and its oracle properties. Journal of the American Statistical Association, 101, 1418–1429.
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 2013 Springer Science+Business Media New York
About this chapter
Cite this chapter
Chakraborty, B., Moodie, E.E.M. (2013). Inference and Non-regularity. In: Statistical Methods for Dynamic Treatment Regimes. Statistics for Biology and Health. Springer, New York, NY. https://doi.org/10.1007/978-1-4614-7428-9_8
Download citation
DOI: https://doi.org/10.1007/978-1-4614-7428-9_8
Published:
Publisher Name: Springer, New York, NY
Print ISBN: 978-1-4614-7427-2
Online ISBN: 978-1-4614-7428-9
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)