Definition of the Subject and Introduction
Statistics may be defined as the study of uncertainty: how to measure it, and how to make choices in the face ofit. Uncertainty is quantified via probability, of which there are two leading paradigms, frequentist (discussed in Sect. “ Comparison with the Frequentist StatisticalParadigm”) and Bayesian. In the Bayesian approach to probability the primitive constructs aretrue-false propositions A whose truth status is uncertain, and the probabilityof A is the numerical weight of evidence in favor of A, constrained toobey a set of axioms to ensure that Bayesian probabilities are coherent (internally logically consistent).
The discipline of statistics may be divided broadly into four activities: description (graphical and numericalsummaries of a data set y, without attempting to reason outward from it; this activity is almost entirelynon‐probabilistic and will not be discussed further here), inference(drawing probabilistic conclusions about...
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Abbreviations
- Bayes' theorem ; prior, likelihood and posterior distributions:
-
Given (a) θ, something of interest which is unknown to the person making an uncertainty assessment, conveniently referred to as You, (b) y, an information source which is relevant to decreasing Your uncertainty about θ, (c) a desire to learn about θ from y in a way that is both internally and externally logically consistent, and (d) \( { \mathcal{B} } \), Your background assumptions and judgments about how the world works, as these assumptions and judgments relate to learning about θ from y, it can be shown that You are compelled in this situation to reason within the standard rules of probability as the basis of Your inferences about θ, predictions of future data \( { y^{*} } \), and decisions in the face of uncertainty (see below for contrasts between inference, prediction and decision‐making), and to quantify Your uncertainty about any unknown quantities through conditional probability distributions. When inferences about θ are the goal, Bayes' Theorem provides a means of combining all relevant information internal and external to y:
$$ p(\theta | y,\mathcal{B})=c\,p(\theta | \mathcal{B})\,l (\theta | y, \mathcal{B})\:. $$(1)Here, for example in the case in which θ is a real‐valued vector of length k, (a) \( { p(\theta | \mathcal{B}) } \) is Your prior distribution about θ given \( { \mathcal{B} } \) (in the form of a probability density function), which quantifies all relevant information available to You about θ external to y, (b) c is a positive normalizing constant, chosen to make the density on the left side of the equation integrate to 1, (c) \( { l(\theta | y, \mathcal{B}) } \) is Your likelihood distribution for θ given y and \( { \mathcal{B} } \), which is defined to be a density‐normalized multiple of Your sampling distribution \( { p (\cdot | \theta, \mathcal{B}) } \) for future data values \( { y^{*} } \) given θ and \( { \mathcal{B} } \), but re‐interpreted as a function of θ for fixed y, and (d) \( { p(\theta | y, \mathcal{B}) } \) is Your posterior distribution about θ given y and \( { \mathcal{B} } \), which summarizes Your current total information about θ and solves the basic inference problem.
- Bayesian parametric and non‐parametric modeling:
-
(1) Following de Finetti [23], a Bayesian statistical model is a joint predictive distribution \( { p(y_1,\dots ,\, y_n) } \) for observable quantities y i that have not yet been observed, and about which You are therefore uncertain. When the y i are real‐valued, often You will not regard them as probabilistically independent (informally, the y i are independent if information about any of them does not help You to predict the others); but it may be possible to identify a parameter vector \( { \theta=(\theta_1,\dots ,\theta_k) } \) such that You would judge the y i conditionally independent given θ, and would therefore be willing to model them via the relation
$$ p(y_1,\dots ,y_n | \theta)=\prod_{i=1}^n p(y_i | \theta)\:. $$(2)When combined with a prior distribution \( { p(\theta) } \) on θ that is appropriate to the context, this is Bayesian parametric modeling, in which \( { p(y_i | \theta) } \) will often have a standard distributional form (such as binomial, Poisson or Gaussian). (2) When a (finite) parameter vector that induces conditional independence cannot be found, if You judge your uncertainty about the real‐valued y i exchangeable (see below), then a representation theorem of de Finetti [21] states informally that all internally logically consistent predictive distributions \( { p(y_1,\dots,\,y_n) } \) can be expressed in a way that is equivalent to the hierarchical model (see below)
$$ \begin{aligned} (F | \mathcal{B}) \,\,\,\,\,\, \sim & \,\,\,p(F | \mathcal{B}) \\ (y_i | F, \mathcal{B}) \,\, \stackrel{\textrm{\tiny IID}}{\sim} & \,\,\,\,\,\,\,\,F\:, \end{aligned} $$(3)where (a) F is the cumulative distribution function (CDF) of the underlying process \( { (y_1 , y_2 ,\,\dots) } \) from which You are willing to regard \( { p (y_1 , \dots , y_n) } \) as (in effect) like a random sample and (b) \( { p (F | \mathcal{B}) } \) is Your prior distribution on the space \( { \mathcal{F} } \) of all CDFs on the real line. This (placing probability distributions on infinite‐dimensional spaces such as \( { \mathcal{F} } \)) is Bayesian non‐parametric modeling, in which priors involving Dirichlet processes and/or Pólya trees (see Sect. “Inference: Parametric and Non‐Parametric Modeling of Count Data”) are often used.
- Exchangeability :
-
A sequence \( { y = (y_1, \ldots,\, y_n) } \) of random variables (for \( { n \ge 1 } \)) is (finitely) exchangeable if the joint probability distribution \( { p (y_1, \dots,\, y_n) } \) of the elements of y is invariant under permutation of the indices \( { (1, \dots,\, n) } \), and a countably infinite sequence \( { (y_1, y_2, \dots) } \) is (infinitely) exchangeable if every finite subsequence is finitely exchangeable.
- Hierarchical modeling :
-
Often Your uncertainty about something unknown to You can be seen to have a nested or hierarchical character. One class of examples arises in cluster sampling in fields such as education and medicine, in which students (level 1) are nested within classrooms (level 2) and patients (level 1) within hospitals (level 2); cluster sampling involves random samples (and therefore uncertainty) at two or more levels in such a data hierarchy (examples of this type of hierarchical modeling are given in Sect. “Strengths and Weaknesses of the Two Approaches”). Another, quite different, class of examples of Bayesian hierarchical modeling is exemplified by equation (3) above, in which is was helpful to decompose Your overall predictive uncertainty about \( { (y_1 , \dots ,\, y_n) } \) into (a) uncertainty about F and then (b) uncertainty about the y i given F (examples of this type of hierarchical modeling appear in Sect. “Inference and Prediction: Binary Outcomes with No Covariates” and “Inference: Parametric and Non‐Parametric Modeling of Count Data”).
- Inference, prediction and decision‐making; samplesand populations:
-
Given a data source y, inference involves drawing probabilistic conclusions about the underlying process that gave rise to y, prediction involves summarizing uncertainty about future observable data values \( { y^{*} } \), and decision‐making involves looking for optimal behavioral choices in the face of uncertainty (about either the underlying process, or the future, or both). In some cases inference takes the form of reasoning backwards from a sample of data values to a population: a (larger) universe of possible data values from which You judge that the sample has been drawn in a manner that is representative (i. e., so that the sampled and unsampled values in the population are (likely to be) similar in relevant ways).
- Mixture modeling:
-
Given y, unknown to You, and \( { \mathcal{B} } \), Your background assumptions and judgments relevant to y, You have a choice: You can either model (Your uncertainty about) y directly, through the probability distribution \( { p(y|\mathcal{B}) } \), or (if that is not feasible) You can identify a quantity x upon which You judge y to depend and model y hierarchically, in two stages: first by modeling x, through the probability distribution \( { p (x | \mathcal{B}) } \), and then by modeling y given x, through the probability distribution \( { p(y|x,\mathcal{B}) } \):
$$ p (y | \mathcal{B}) = \int_\mathcal{X} p (y | x , \mathcal{B}) \, p (x | \mathcal{B}) \, \text{d} x\:, $$(4)where \( { \mathcal{X} } \) is the space of possible values of x over which Your uncertainty is expressed. This is mixture modeling , a special case of hierarchical modeling (see above). In hierarchical notation (4) can be re‐expressed as
$$ y = \left\{\begin{array}{c} x \\ (y | x) \end{array}\right\}\:. $$(5)Examples of mixture modeling in this article include (a) equation (3) above, with F playing the role of x; (b) the basic equation governing Bayesian prediction, discussed in Sect. “The Bayesian Statistical Paradigm”; (c) Bayesian model averaging (Sect. “The Bayesian Statistical Paradigm”); (d) de Finetti's representation theorem for binary outcomes (Sect. “Inference and Prediction: Binary Outcomes with No Covariates”); (e) random‐effects parametric and non‐parametric modeling of count data (Sect. “Inference: Parametric and Non‐Parametric Modeling of Count Data”); and (f) integrated likelihoods in Bayes factors (Sect. “Decision‐Making: Variable Selection in Generalized Linear Models; Bayesian Model Selection”).
- Probability – frequentist and Bayesian:
-
In the frequentist probability paradigm, attention is restricted to phenomena that are inherently repeatable under (essentially) identical conditions; then, for an event A of interest, \( { P_f (A) } \) is the limiting relative frequency with which A occurs in the (hypothetical) repetitions, as the number of repetitions \( { n \rightarrow \infty }\). By contrast, YourBayesian probability \( { P_B (A|\mathcal{B}) } \) is the numerical weightof evidence, given Your background information \( { \mathcal{B} } \) relevantto A, in favor of a true-false proposition A whose truth status is uncertain to You, obeying a series of reasonable axioms to ensure that Your Bayesian probabilities are internally logically consistent.
- Utility:
-
To ensure internal logical consistency, optimal decision‐making proceeds by (a) specifying a utility function \( { U (a , \theta_0) } \) quantifying the numerical value associated with taking action a if the unknown is really θ0 and (b) maximizing expected utility, where the expectation is taken over uncertainty in θ as quantified by the posterior distribution \( { p (\theta | y , \mathcal{B}) } \).
Bibliography
Abdellaoui M (2000) Parameter‐free elicitation of utility and probability weighting functions. Manag Sci 46:1497–1512
Aleskerov F, Bouyssou D, Monjardet B (2007) Utility Maximization, Choice and Preference, 2nd edn. Springer, New York
Arrow KJ (1963) Social Choice and Individual Values, 2nd edn. Yale University Press, New Haven CT
Barlow RE, Wu AS (1981) Preposterior analysis of Bayes estimators of mean life. Biometrika 68:403–410
Bayes T (1764) An essay towards solving a problem in the doctrine of chances. Philos Trans Royal Soc Lond 53:370–418
Berger JO (1985) Statistical Decision Theory and Bayesian Analysis. Springer, New York
Berger JO (2006) The case for objective Bayesian analysis (with discussion). Bayesian Anal 1:385–472
Berger JO, Betro B, Moreno E, Pericchi LR, Ruggeri F, Salinetti G, Wasserman L (eds) (1995) Bayesian Robustness. Institute of Mathematical Statistics Lecture Notes‐Monograph Series, vol 29. IMS, Hayward CA
Berger JO, Pericchi LR (2001) Objective Bayesian methods for model selection: introduction and comparison. In: Lahiri P (ed) Model Selection. Monograph Series, vol 38. Institute of Mathematical Statistics Lecture Notes Series, Beachwood, pp 135–207
Bernardo JM (1979) Reference posterior distributions for Bayesian inference (with discussion). J Royal Stat Soc, Series B 41:113–147
Bernardo JM, Smith AFM (1994) Bayesian Theory. Wiley, New York
Blavatskyy P (2006) Error propagation in the elicitation of utility and probability weighting functions. Theory Decis 60:315–334
Brown PJ, Vannucci M, Fearn T (1998) Multivariate Bayesian variable selection and prediction. J Royal Stat Soc, Series B 60:627–641
Browne WJ, Draper D (2006) A comparison of Bayesian and likelihood methods for fitting multilevel models (with discussion). Bayesian Anal 1:473–550
Buck C, Blackwell P (2008) Bayesian construction of radiocarbon calibration curves (with discussion). In: Case Studies in Bayesian Statistics, vol 9. Springer, New York
Chipman H, George EI, McCulloch RE (1998) Bayesian CART model search (with discussion). J Am Stat Assoc 93:935–960
Christiansen CL, Morris CN (1997) Hierarchical Poisson regression modeling. J Am Stat Assoc 92:618–632
Clyde M, George EI (2004) Model uncertainty. Stat Sci 19:81–94
Das S, Chen MH, Kim S, Warren N (2008) A Bayesian structural equations model for multilevel data with missing responses and missing covariates. Bayesian Anal 3:197–224
de Finetti B (1930) Funzione caratteristica di un fenomeno aleatorio. Mem R Accad Lincei 4:86–133
de Finetti B (1937). La prévision: ses lois logiques, ses sources subjectives. Ann Inst H Poincaré 7:1–68 (reprinted in translation as de Finetti B (1980) Foresight: its logical laws, its subjective sources. In: Kyburg HE, Smokler HE (eds) Studies in Subjective Probability. Dover, New York, pp 93–158)
de Finetti B (1938/1980). Sur la condition d'équivalence partielle. Actual Sci Ind 739 (reprinted in translation as de Finetti B (1980) On the condition of partial exchangeability. In: Jeffrey R (ed) Studies in Inductive Logic and Probability. University of California Press, Berkeley, pp 193–206)
de Finetti B (1970) Teoria delle Probabilità, vol 1 and 2. Eunaudi, Torino (reprinted in translation as de Finetti B (1974–75) Theory of probability, vol 1 and 2. Wiley, Chichester)
Dey D, Müller P, Sinha D (eds) (1998) Practical Nonparametric and Semiparametric Bayesian Statistics. Springer, New York
Draper D (1995) Assessment and propagation of model uncertainty (with discussion). J Royal Stat Soc, Series B 57:45–97
Draper D (1999) Model uncertainty yes, discrete model averaging maybe. Comment on: Hoeting JA, Madigan D, Raftery AE, VolinskyCT (eds) Bayesian model averaging: a tutorial. Stat Sci14:405–409
Draper D (2007) Bayesian multilevel analysis and MCMC. In: de Leeuw J, Meijer E (eds) Handbook of Multilevel Analysis. Springer, New York, pp 31–94
Draper D, Hodges J, Mallows C, Pregibon D (1993) Exchangeability and data analysis (with discussion). J Royal Stat Soc, Series A 156:9–37
Draper D, Krnjajić M (2008) Bayesian model specification. Submitted
Duran BS, Booker JM (1988) A Bayes sensitivity analysis when using the Beta distribution as a prior. IEEE Trans Reliab 37:239–247
Efron B (1979) Bootstrap methods. Ann Stat 7:1–26
Ferguson T (1973) A Bayesian analysis of some nonparametric problems. Ann Stat 1:209–230
Ferreira MAR, Lee HKH (2007) Multiscale Modeling. Springer, New York
Fienberg SE (2006) When did Bayesian inference become “Bayesian”? Bayesian Anal 1:1–40
Fishburn PC (1970) Utility Theory for Decision Making. Wiley, New York
Fishburn PC (1981) Subjective expected utility: a review of normative theories. Theory Decis 13:139–199
Fisher RA (1922) On the mathematical foundations of theoretical statistics. Philos Trans Royal Soc Lond, Series A 222:309–368
Fisher RA (1925) Statistical Methods for Research Workers. Oliver and Boyd, Edinburgh
Fouskakis D, Draper D (2008) Comparing stochastic optimization methods for variable selection in binary outcome prediction, with application to health policy. J Am Stat Assoc, forthcoming
Gauss CF (1809) Theoria Motus Corporum Coelestium in Sectionibus Conicis Solem Ambientium, vol 2. Perthes and Besser, Hamburg
Gelfand AE, Smith AFM (1990) Sampling‐based approaches to calculating marginal densities. J Am Stat Assoc 85:398–409
Gelman A, Meng X-L (2004) Applied Bayesian Modeling and Causal Inference From Incomplete‐Data Perspectives. Wiley, New York
George EI, Foster DP (2000) Calibration and empirical Bayes variable selection. Biometrika 87:731–747
Gilks WR, Richardson S, Spiegelhalter DJ (eds) (1996) Markov Chain Monte Carlo in Practice. Chapman, New York
Goldstein M (2006) Subjective Bayesian analysis: principles and practice (with discussion). Bayesian Anal 1:385–472
Good IJ (1950) Probability and the Weighing of Evidence. Charles Griffin, London
Green P (1995) Reversible jump Markov chain Monte carlo computation and Bayesian model determination. Biometrika 82:711–713
Hacking I (1984) The Emergence of Probability. University Press, Cambridge
Hanson TE, Kottas A, Branscum AJ (2008) Modelling stochastic order in the analysis of receiver operating characteristic data: Bayesian non‐parametric approaches. J Royal Stat Soc, Series C (Applied Statistics) 57:207–226
Hellwig K, Speckbacher G, Weniges P (2000) Utility maximization under capital growth constraints. J Math Econ 33:1–12
Hendriksen C, Lund E, Stromgard E (1984) Consequences of assessment and intervention among elderly people: a three year randomized controlled trial. Br Med J 289:1522–1524
Hoeting JA, Madigan D, Raftery AE, Volinsky CT (1999) Bayesian model averaging: a tutorial. Stat Sci 14:382–417
Jaynes ET (2003) Probability Theory: The Logic of Science. Cambridge University Press, Cambridge
Jeffreys H (1931) Scientific Inference. Cambridge University Press, Cambridge
Jordan MI, Ghahramani Z, Jaakkola TS, Saul L (1999) An introduction to variational methods for graphical models. Mach Learn 37:183–233
Kadane JB (ed) (1996) Bayesian Methods and Ethics in a Clinical Trial Design. Wiley, New York
Kadane JB, Dickey JM (1980) Bayesian decision theory and the simplification of models. In: Kmenta J, Ramsey J (eds) Evaluation of Econometric Models. Academic Press, New York
Kahn K, Rubenstein L, Draper D, Kosecoff J, Rogers W, Keeler E, Brook R (1990) The effects of the DRG-based Prospective Payment System on quality of care for hospitalized Medicare patients: An introduction to the series (with discussion). J Am Med Assoc 264:1953–1955
Kass RE, Raftery AE (1995) Bayes factors. J Am Stat Assoc 90:773–795
Kass RE, Wasserman L (1996) The selection of prior distributions by formal rules. J Am Stat Assoc 91:1343–1370
Key J, Pericchi LR, Smith AFM (1999) Bayesian model choice: what and why? (with discussion). In: Bernardo JM, Berger JO, Dawid AP, Smith AFM (eds) Bayesian Statistics 6. Clarendon Press, Oxford, pp 343–370
Krnjajić M, Kottas A, Draper D (2008) Parametric and nonparametric Bayesian model specification: a case study involving models for count data. Comput Stat Data Anal 52:2110–2128
Laplace PS (1774) Mémoire sur la probabilité des causes par les évenements. Mém Acad Sci Paris 6:621–656
Laplace PS (1812) Théorie Analytique des Probabilités. Courcier, Paris
Laud PW, Ibrahim JG (1995) Predictive model selection. J Royal Stat Soc, Series B 57:247–262
Lavine M (1992) Some aspects of Pólya tree distributions for statistical modelling. Ann Stat 20:1222–1235
Leamer EE (1978) Specification searches: Ad hoc inference with non‐experimental data. Wiley, New York
Leonard T, Hsu JSJ (1999) Bayesian Methods: An Analysis for Statisticians and Interdisciplinary Researchers. Cambridge University Press, Cambridge
Lindley DV (1965) Introduction to Probability and Statistics. Cambridge University Press, Cambridge
Lindley DV (1968) The choice of variables in multiple regression (with discussion). J Royal Stat Soc, Series B 30:31–66
Lindley DV (2006) Understanding Uncertainty. Wiley, New York
Lindley DV, Novick MR (1981) The role of exchangeability in inference. Ann Stat 9:45–58
Little RJA (2006) Calibrated Bayes: A Bayes/frequentist roadmap. Am Stat 60:213–223
Mangel M, Munch SB (2003) Opportunities for Bayesian analysis in the search for sustainable fisheries. ISBA Bulletin 10:3–5
McCullagh P, Nelder JA (1989) Generalized Linear Models, 2nd edn. Chapman, New York
Meng XL, van Dyk DA (1997) The EM Algorithm: an old folk song sung to a fast new tune (with discussion). J Royal Stat Soc, Series B 59:511–567
Merl D, Prado R (2007) Detecting selection in DNA sequences: Bayesian modelling and inference (with discussion). In: Bernardo JM, Bayarri MJ, Berger JO, Dawid AP, Heckerman D, Smith AFM, West M (eds) Bayesian Statistics 8. University Press, Oxford, pp 1–22
Metropolis N, Rosenbluth AW, Rosenbluth MN, Teller AH, Teller E (1953) Equations of state calculations by fast computing machine. J Chem Phys 21:1087–1091
Morris CN, Hill J (2000) The Health Insurance Experiment: design using the Finite Selection Model. In: Morton SC, Rolph JE (eds) Public Policy and Statistics: Case Studies from RAND. Springer, New York, pp 29–53
Müller P, Quintana F (2004) Nonparametric Bayesian data analysis. Stat Sci 19:95–110
Müller P, Quintana F, Rosner G (2004) Hierarchical meta‐analysis over related non‐parametric Bayesian models. J Royal Stat Soc, Series B 66:735–749
Munch SB, Kottas A, Mangel M (2005) Bayesian nonparametric analysis of stock‐recruitment relationships. Can J Fish Aquat Sci 62:1808–1821
Neyman J (1937) Outline of a theory of statistical estimation based on the classical theory of probability. Philos Trans Royal Soc Lond A 236:333–380
Neyman J, Pearson ES (1928) On the use and interpretation of certain test criteria for purposes of statistical inference. Biometrika 20:175–240
O'Hagan A, Buck CE, Daneshkhah A, Eiser JR, Garthwaite PH, Jenkinson DJ, Oakley JE, Rakow T (2006) Uncertain Judgements. Eliciting Experts' Probabilities. Wiley, New York
O'Hagan A, Forster J (2004) Bayesian Inference, 2nd edn. In: Kendall's Advanced Theory of Statistics, vol 2B. Arnold, London
Parmigiani G (2002) Modeling in medical decision‐making: A Bayesian approach. Wiley, New York
Pearson KP (1895) Mathematical contributions to the theory of evolution, II. Skew variation in homogeneous material. Proc Royal Soc Lond 57:257–260
Pebley AR, Goldman N (1992) Family, community, ethnic identity, and the use of formal health care services in Guatemala. Working Paper 92-12. Office of Population Research, Princeton
Pettit LI (1990) The conditional predictive ordinate for the Normal distribution. J Royal Stat Soc, Series B 52:175–184
Pérez JM, Berger JO (2002) Expected posterior prior distributions for model selection. Biometrika 89:491–512
Polson NG, Stroud JR, Müller P (2008) Practical filtering with sequential parameter learning. J Royal Stat Soc, Series B 70:413–428
Rashbash J, Steele F, Browne WJ, Prosser B (2005) A User's Guide to MLwiN, Version 2.0. Centre for Multilevel Modelling, University of Bristol, Bristol UK; available at www.cmm.bristol.ac.uk Accessed 15 Aug 2008
Richardson S, Green PJ (1997) On Bayesian analysis of mixtures with an unknown number of components (with discussion). J Royal Stat Soc, Series B 59:731–792
Rios Insua D, Ruggeri F (eds) (2000) Robust Bayesian Analysis. Springer, New York
Rodríguez A, Dunston DB, Gelfand AE (2008) The nested Dirichlet process. J Am Stat Assoc, 103, forthcoming
Rodríguez G, Goldman N (1995) An assessment of estimation procedures for multilevel models with binary responses. J Royal Stat Soc, Series A 158:73–89
Rubin DB (1984) Bayesianly justifiable and relevant frequency calculations for the applied statistician. Ann Stat 12:1151–1172
Rubin DB (2005) Bayesian inference for causal effects. In: Rao CR, Dey DK (eds) Handbook of Statistics: Bayesian Thinking, Modeling and Computation, vol 25. Elsevier, Amsterdam, pp 1–16
Sabatti C, Lange K (2008) Bayesian Gaussian mixture models for high‐density genotyping arrays. J Am Stat Assoc 103:89–100
Sansó B, Forest CE, Zantedeschi D (2008) Inferring climate system properties using a computer model (with discussion). Bayesian Anal 3:1–62
Savage LJ (1954) The Foundations of Statistics. Wiley, New York
Schervish MJ, Seidenfeld T, Kadane JB (1990) State‐dependent utilities. J Am Stat Assoc 85:840–847
Seidou O, Asselin JJ, Ouarda TMBJ (2007) Bayesian multivariate linear regression with application to change point models in hydrometeorological variables. In: Water Resources Research 43, W08401, doi:10.1029/2005WR004835.
Spiegelhalter DJ, Abrams KR, Myles JP (2004) Bayesian Approaches to Clinical Trials and Health‐Care Evaluation. Wiley, New York
Spiegelhalter DJ, Best NG, Carlin BP, van der Linde A (2002) Bayesian measures of model complexity and fit (with discussion). J Royal Stat Soc, Series B 64:583–640
Spiegelhalter DJ, Thomas A, Best NG (1999) WinBUGS Version 1.2 User Manual. MRC Biostatistics Unit, Cambridge
Stephens M (2000) Bayesian analysis of mixture models with an unknown number of components – an alternative to reversible‐jump methods. Ann Stat 28:40–74
Stigler SM (1986) The History of Statistics: The Measurement of Uncertainty Before 1900. Harvard University Press, Cambridge
Stone M (1974) Cross‐validation choice and assessment of statistical predictions (with discussion). J Royal Stat Soc, Series B 36:111–147
Wald A (1950) Statistical Decision Functions. Wiley, New York
Weerahandi S, Zidek JV (1981) Multi‐Bayesian statistical decision theory. J Royal Stat Soc, Series A 144:85–93
Weisberg S (2005) Applied Linear Regression, 3rd edn. Wiley, New York
West M (2003) Bayesian factor regression models in the “large p, small n paradigm.” Bayesian Statistics 7:723–732
West M, Harrison PJ (1997) Bayesian Forecasting and Dynamic Models. Springer, New York
Whitehead J (2006) Using Bayesian decision theory in dose‐escalation studies. In: Chevret S (ed) Statistical Methods for Dose‐Finding Experiments. Wiley, New York, pp 149–171
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2009 Springer-Verlag
About this entry
Cite this entry
Draper, D. (2009). Bayesian Statistics. In: Meyers, R. (eds) Encyclopedia of Complexity and Systems Science. Springer, New York, NY. https://doi.org/10.1007/978-0-387-30440-3_31
Download citation
DOI: https://doi.org/10.1007/978-0-387-30440-3_31
Publisher Name: Springer, New York, NY
Print ISBN: 978-0-387-75888-6
Online ISBN: 978-0-387-30440-3
eBook Packages: Physics and AstronomyReference Module Physical and Materials ScienceReference Module Chemistry, Materials and Physics