Abstract
We have theories of statistics and we use statistical methods to solve subject matter problems. One might expect that the theories would affect how the methods are used, but they do so only superficially, because all statistical theories are quite incomplete as descriptions of and prescriptions for statistical practice. This paper sketches an extension of Bayesian theory that might address this incompleteness. The extension is based on five ideas:
-
1.
The product of a statistical analysis is an argument — not an HPD region, posterior distribution, decision, or other data summary, but the entire argument, including premises and logical steps.
-
2.
Arguments come in several logically distinct types, with an argument’s type being defined by the form of its conclusion. The paper catalogs the types of argument and identifies the main burden of each.
-
3.
EDA and model-building activities establish a plausible, tractable baseline argument for a given problem and dataset.
-
4.
Diagnostics and sensitivity analyses vary the premises of the baseline argument and display the resulting variation in the conclusion (as opposed to intermediate quantities).
-
5.
An argument is strong to the extent that:
-
its premises are conclusions of strong arguments, or
-
the region of premises yielding the same conclusion is large.
A simple example demonstrates the mechanics of the extended theory. The example is then generalized to draw implications for statistical foundations, methods, and computing. This paper amounts to a research agenda, so it poses more problems than it solves.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
Smith, A.F.M. ‘Some Bayesian thoughts on modelling and model choice’, The Statistician, 35, 97–102 (1986)
Hill, BM ‘A theory of Bayesian data analysis’, In Bayesian and Likelihood Methods in Statistics and Econometrics: Essays in Honor of George A. Barnard, eds. S Geisser, JS Hodges, SJ Press, A Zellner, Amsterdam: North-Holland, 49–74 (1990)
Berger, JO ‘Contributed discussion’, In Case Studies in Bayesian Statistics, eds. C Gatsonis, JS Hodges, RE Kass, ND Singpurwalla. New York: Springer-Verlag, 302–303 (1993)
West, M. and Harrison, J. Bayesian Forecasting and Dynamic Models. New York: Springer-Verlag (1989)
Fisher, R.A. Statistical Methods and Scientific Inference, 3rd edition, New York: Hafner (1973)
Efron, B. ‘Why isn’t everyone a Bayesian?’, American Statistician, 40, 1–11 1986
Fischl, MA, Richman, DD, Grieco MH, et al. ‘The efficacy of azi- dothymidine (AZT) in the treatment of patients with AIDS and AIDS- related complex’, New England Journal of Medicine, 317, 185–191, (1987)
Kahn, J.O., Lagakos, S.W., Richman, D.D., et al. ‘A controlled trial comparing continued zidovudine [AZT] with didanosine in human immunodeficiency virus infection’, New England Journal of Medicine, 337, 581–587, (1992)
Litterman, R.B. ‘A statistical approach to economic forecasting’, Journal of Business and Economic Statistics, 4, 1–4, (1986)
Mallows, CL ‘Data description’, In Scientific Inference, Data Analysis, and Robustness, eds. GEP Box, T Leonard, C-F Wu, Academic Press, 135–152 (1983)
Draper, DC, Hodges, JS, Mallows, CL, and Pregibon, D ‘Exchangeability and data analysis (with discussion)’. Journal of the Royal Statistical Society, Series A, 156, 9–37 (1993)
Carter, H.R., Page, G.W., and Ford, R.G. ‘The importance of rehabilitation center data in determining the impacts of the 1986 oil spill on marine birds in central California’, Wildlife Journal, 10, 9–14 (1987)
Blakeman, B.M., Pifarre, R., Sullivan H., et al. ‘High-risk heart surgery in the heart transplant candidate’, Journal of Heart Transplantation, 5, 468–472 (1990)
Tukey, J W. ‘Some thoughts on clinical trials, especially problems of multiplicity’, Science, 198, 679–684 (1977)
Learner, E.E. ‘False models and post-data model construction’, Journal of the American Statistical Association, 69, 122–131 (1974)
Lindley, D.V. Discussion of Efron [6]
Lindley, D.V. ‘The 1988 Wald Memorial Lectures: The present position in Bayesian statistics (with discussion)’, Statistical Science, 5, 44–89 (1990)
Berry, D.A. ‘Subgroup analyses’, Biometrics, 47, 1227–1230 (1990)
Neaton, J. and Went worth, D. “Relationship of serum cholesterol and blood pressure measured prior to HIV-infection with risk of death from AIDS”, Unpublished manuscript, Division of Biostatistics, School of Public Health, University of Minnesota.
Fleming, TR ‘Surrogate markers in AIDS and cancer trials’, Statistics in Medicine, in press (1995)
De Gruttola, V., Wulfsohn, M., Fischl, M.A., and Tsiatis, A.A. ‘Modeling the relationship between survival and CD4 lymphocytes in patients with AIDS and AIDS-related complex’, Journal of Acquired Immune Deficiency Syndromes, 6, 359–365 (1993)
Choi, S, Lagakos, SW, Schooley, RT, and Volberding, PA ‘CD4+ lymphocytes are an incomplete surrogate marker for clinical progression in persons with asymptomatic HIV infection taking zidovudine’. Annals of Internal Medicine, 118, 674–680 (1993)
Schmitz, J, “Massive Marketing Datasets”, presented on 7 July 1995 at “Statistical Challenges and Possible Approaches in the Analysis of Massive Data Sets,” a conference sponsored by the Committee on Applied and Theoretical Statistics; Washington, DC.
Koech, D., et al. ‘Low-dose oral alpha-interferon therapy for patients seropositive for human immunodeficiency virus type-1 (HIV-1)’, Molecular Biotherapy, 2, 91–95 (1990)
Obel, A.O. and Koech, D. ‘Outcome of intervention with or without low- dose oral alpha-interferon in 32 HIV-1 seropositive patients in a referral hospital’. East African Medical Journal, 67 (7), 71–76 (1990)
Hardy, W.D., Feinberg, J., Finkelstein, D.M., et al. ‘A controlled trial of trimethprim-sulfamethoxazole or aerosolized pentamidine for secondary prophylaxis of Pneumocystis carinii pneumonia in patients with the acquired immuneodeficiency syndrome’, New England Journal of Medicine, 327, 1842–1848 (1992)
Schneider, M.M.E., Hoepelman, A.I.M., Schattenkerk, J.K.M.E, et al. ‘A controlled trial of aerosolized pentamidine or trimethoprim-sulfamethoxazole as primary prophylaxis against Pneumocystis carinii pneumonia in patients with human immunodeficiency virus infection’, New England Journal of Medicine, 327, 1836–1841 (1992)
Cook, R.D. and Weisberg, S. Residuals and Influence in Regression, New York: Chapman and Hall. (1982)
Cook, RD ‘Assessment of local influence (with discussion)’, Journal of the Royal Statistical Society, SeriesB, 48, 133–169 (1986)
Johnson, W.O. and Geisser, S. ‘A predictive view of the detection and characterization of influential observations in regression analysis’, Journal of the American Statistical Assn, 78, 137–144 (1983)
McCulloch, R.E. ‘Local model influence’, Journal of the American Statistical Assn, 84, 473–478 (1989)
Carlin, BP and Poison, NG ‘Monte Carlo Bayesian methods for discrete regression models and categorical time series’, In Bayesian Statistics 4, eds. JM Bernardo, JO Berger, AP Dawid, AFM Smith, Oxford University Press, 577–586, (1992)
Kass, R.E., Tierney, L., and Kadane, J.B. ‘Approximate methods for assessing influence and sensitivity in Bayesian analysis’, Biometrika, 76, 663–674, (1989)
Gelfand, AE, Dey, DK, and Chang, H ‘Model determination using predictive distributions with implementation via sampling-based methods’, In Bayesian Statistics 4, eds. JM Bernardo, JO Berger, AP Dawid, AFM Smith, Oxford University Press, 147–167, (1992)
Draper, D., Hodges, J.S., Mallows, C.L., and Pregibon, D. ‘Exchangeability and data analysis (with discussion)’, Journal of the Royal Statistical Society, Series A, 156, 9–37 (1993)
Carlin, B.P., Kass, R.E., Lerch, F.J., and Huguenard, B.R. ‘Predicting working memory failure: A subjective Bayesian approach to model selection’, Journal of the American Statistical Assn, 87, 319–327 (1992)
Gibbons, R.D., Hedeker, D., Charles, S.C., and Frisch, P. ‘A random- effects probit model for predicting medical malpractice claims’, Journal of the American Statistical Assn, 89, 760–767 (1994)
Caulkins, J.P. and Padman, R. ‘Quantity discounts and quality premia for illicit drugs’, Journal of the American Statistical Assn, 88, 748–757 (1993)
Zaslavsky, A.M. ‘Combining census, dual-system, and evaluation study data to estimate population shares’, Journal of the American Statistical Assn, 88, 1092–1105 (1993)
Crawford, SL, Johnson, WG, and Laird, NM ‘Bayes analysis of model- based methods for nonignorable nonresponse in the Harvard Medical Practice Survey’, In Case Studies in Bayesian Statistics, eds. C Gatsonis, JS Hodges, RE Kass, ND Singpurwalla, New York: Springer-Verlag, 78–117 (1993)
Ehrenberg, ASC and Bound, JA ‘Predictability and prediction (with discussion)’, Journal of the Royal Statistical Society, Series A, 156, 167–206 (1993)
Carlin, B.P, and Louis, T.A. ‘Identifying prior distributions that produce specific decisions, with application to monitoring clinical trials.’, In Bayesian Analysis of Statistics and Econometrics: Essays in Honor of Arnold Zellner, eds. D.A. Berry, K.M. Chaloner, J.K. Geweke, New York: Wiley. (1996)
Carlin, B.P., Chaloner, K.M., Louis, T.A., and Rhame, F.S. ‘Elicitation, monitoring, and analysis for an AIDS clinical trial’, In Case Studies in Bayesian StatisticsVolume II, eds. C. Gatsonis, J.S. Hodges, R.E. Kass, N.D. Singpurwalla, New York: Springer-Verlag (1995)
Freedman, L.S. and Spiegelhalter, D.J. ‘Application of Bayesian statistics to decision making during a clinical trial’, Statistics in Medicine, 11, 23–35 (1992)
Sargent, D. and Carlin, B.P ‘Robust Bayesian design and analysis of clinical trials via prior partitioning (with discussion)’, Research Report 94–016, Division of Biostatistics, University of Minnesota, 1994. To appear in the IMS Lecture Note Series.
Hodges, J.S. ‘Uncertainty, policy analysis, and statistics (with discussion)’, Statistical Science, 2, 259–291 (1987)
Hodges, J.S. ‘Six (or so) things you can do with a bad model’, Operations Research, 39, 355–365 (1991)
Hodges, J.S. and Dewar, J.A. ‘Is it you or your model talking? A framework for model validation’, RAND, R-4114-AF/A/OSD, Santa Monica, California (1992)
Bankes, SC ’Exploratory modeling and the use of simulation for policy analysis’, RAND, N-3093-A, Santa Monica, California (1992)
Dewar, J.A., Bankes, S.C., Hodges, JS., et al. ‘Credible uses of the Distributed Interactive Simulation (DIS) System’ RAND, MR-607-A, Santa Monica, California (1995)
Smith, AFM ‘Present position and potential developments: Some personal views of Bayesian statistics (with discussion)’, Journal of the Royal Statistical Society, Series A, 147, 245–259 (1984)
Cox, D.R. and Snell, E.J. Applied Statistics: Principles and Examples. London: Chapman and Hall (1981)
Weisberg, S. ’Some principles for regression diagnostics and influence analysis: Comments on “Developments in linear regression methodology: 1959–1982” by RR Hocking’, Technometrics, 25, 240–244 (1983)
Cox, D.R. ‘Nonlinear models, residuals, and transformations’, Math. Operations forsch. Statist. Ser. Statistics, 8, 3–22 (1977)
Ramsay, J.O. and Novick, M.R. ‘PLU robust Bayesian decision theory: Point estimation’, Journal of the American Statistical Association, 75, 901–907 (1980)
Wasserman, L ‘Recent methodological advances in robust Bayesian inference’, In Bayesian Statistics 4, eds. JM Bernardo, JO Berger, AP Dawid, AFM Smith, Oxford University Press, 483–502 (1992)
Kadane, JB and Schum, DA ‘Opinions in dispute: The Sacco-Vanzetti case’, In Bayesian Statistics 4, eds. JM Bernardo, JO Berger, AP Dawid, AFM Smith, Oxford University Press, 267–287 (1992)
Lindley, D.V. and Singpurwalla, N.D. 4On the evidence needed to reach agreed action between adversaries, with application to acceptance sampling’, Journal of the American Statistical Association, 86, 933–937 (1991)
Rosenbaum, P.R. ‘Sensitivity analysis for certain permutation inferences in matched observational studies’, Biometrika, 74, 13–26 (1987)
Learner, EE Specification Searches, New York: Wiley (1978)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1996 Springer Science+Business Media New York
About this chapter
Cite this chapter
Hodges, J.S. (1996). Statistical Practice as Argumentation: A Sketch of a Theory of Applied Statistics. In: Lee, J.C., Johnson, W.O., Zellner, A. (eds) Modelling and Prediction Honoring Seymour Geisser. Springer, New York, NY. https://doi.org/10.1007/978-1-4612-2414-3_2
Download citation
DOI: https://doi.org/10.1007/978-1-4612-2414-3_2
Publisher Name: Springer, New York, NY
Print ISBN: 978-1-4612-7529-9
Online ISBN: 978-1-4612-2414-3
eBook Packages: Springer Book Archive