DOCSLIB.ORG

Dicing with the Unknown

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Dicing with the Unknown focus DDicingicing wwithith tthehe uunknownnknown There are many things that I am uncertain about, says Tony O’Hagan. Some are merely unknown to me, while others are unknowable. This article is about different kinds of uncertainty, and how the distinction between them impinges on the foundations of Probability and Statistics. Two kinds of uncertainty certainties are potentially reducible by further investigation. But it is easy to see how much There are things that I am uncertain about sim- more fundamental it should be for statisticians, ply because I lack knowledge, and in principle for whom randomness and uncertainty are their my uncertainty might be reduced by gathering very raison d’être. more information. Others are subject to random The theory of Statistics rests on describing variability, which is unpredictable no matter how uncertainties by using probability. A probability much information I might get; these are the un- near 1 represents an event that is almost certain knowables. The two kinds of uncertainty have to occur, while a probability near 0 represents been debated by philosophers, who have given one that is almost certain not to occur. As we them the names epistemic uncertainty (due to move away from these extremes towards the lack of knowledge) and aleatory uncertainty (due probability of ½, there is increasing uncertainty. to randomness). Here, however, is where we meet the fundamental Examples of aleatory uncertainty are famil- dichotomy between the two principal theories of iar to students of probability theory, and include Statistics: the frequentist and Bayesian theories. the outcomes of tossing dice and drawing cards One characterisation of the difference between from a shuffl ed pack. In statistics, aleatory un- these two schools of statistical theory is that fre- certainty is present in almost all data that we Charles Trevelyan quentists do not accept that aleatory uncertainty obtain, due to random variability between the can be described or measured by probabilities, members of a population that we sample from, while Bayesians are happy to use probabilities to or to random measurement errors. quantify any kind of uncertainty. Examples of epistemic uncertainty are all around us. I am uncertain about the atomic weight of zinc, about the population of the city Two kinds of probability of Paris, and about whether the river Thames froze over in London during the winter of Delving more deeply, the root of this disagree- 1600–1601. At least for the fi rst two of these, ment lies in what probability means. Almost my uncertainty could be resolved by looking in everyone who encounters probability for the fi rst a suitable reference book. It may be that there time in their education will be taught it using is no such source of information about the river aleatory uncertainties, like the familiar games of Thames in 1600–1601, but in principle this is tossing dice or coins. And the way probability will a question that might be resolved by histori- be taught is that it represents the long-run fre- cal investigation. Epistemic uncertainty about quency with which the event in question occurs any given question varies from one person to if it is repeated an indefi nite number of times, another. For instance, I have negligible uncer- and this is accordingly known as the frequency tainty about my height. Someone who has seen defi nition of probability. The nature of random me might be able to guess reasonably accurately events is that they are, at least conceptually, re- but could not be certain. Someone who knows peatable in this way. Epistemic uncertainties are only that I am a British male would have more generally not. uncertainty. If probability is to encompass epistemic un- The distinction between aleatory and epis- certainty it needs another defi nition. temic uncertainties is valuable in many areas In Bayesian statistics the probability of a where it is important to appreciate which un- proposition simply represents a degree of belief 132 september2004 112-focus.indd2-focus.indd 113232 112/08/20042/08/2004 115:30:505:30:50 PProcessrocess CCyanyan PProcessrocess MMagentaagenta PProcessrocess YYellowellow PProcessrocess BBlacklack PPANTONEANTONE 11807807 C in the truth of that proposition. Notice in pass- interval were computed from each of the same ing that we tend to use the word “proposition” infi nite sequence of data sets, then 95% of those Six of one and half a (a statement that is either true or false) rather intervals will contain the true value of the pa- than “event” (something which may or may not rameter. dozen of the other occur) when discussing this kind of probability, Neither of them says anything about the If I toss an ordinary coin, my probabil- since “event” has connotations of randomness chance that the null hypothesis is true, or that ity that it will land Heads is 0.5. Suppose and repeatability. A proposition might simply as- the parameter lies in the interval, for these data. now that I have a bag of poker chips, and sert that an event will occur, but it may also refer If we condition on the single set of data in front I know only that some are red and some to a statement with epistemic uncertainty. of us, there is no randomness in the problem, and are green. I have no reason to suppose that The degree-of-belief interpretation of prob- so no frequentist probabilities can be stated. there are equal numbers of red and green ability is sometimes referred to as personal prob- In contrast, Bayesian inference does make chips. Indeed, almost certainly one colour ability or subjective probability because, as noted probability statements about parameters. It can will be more abundant than the other, but already, different people may have different de- do so because the epistemic uncertainty in pa- I have no idea which colour that will be, grees of uncertainty about a proposition. It is rameters can be quantifi ed using the Bayesian’s or how much more abundant it will be than the “subjectivity” of this approach to probabil- personal probability. Indeed, Bayesian inference the other colour. If one chip is to be pulled ity that is most objected to by followers of the describes how the acquisition of data modifi es out of the bag my probability that it will be frequentist theory. Bayesians steadfastly defend (and usually reduces) the uncertainty about red is 0.5. this defi nition, and question the extent to which a parameter, from “prior” uncertainty to “pos- Now surely my uncertainties in the coin their methods are any more subjective than fre- terior” uncertainty. The Bayesian equivalent of toss and the poker chip draw are different— quentist practice, or indeed scientifi c practice a signifi cance test asserts the probability that the coin toss being very familiar and the generally. However, that debate is beyond the the null hypothesis is true. In the same way, the bag of poker chips full of uncertainty—so scope of this article! Bayesian analogue of the confi dence interval why do I give both events the same prob- (usually called a credible interval) has exactly the ability? It is true that the uncertainties interpretation that is so often erroneously attrib- Two kinds of statistics are different. The uncertainty about the uted to the frequentist confi dence interval. coin toss is purely aleatory, whereas there To see the implications of the frequentist po- is clearly epistemic uncertainty about the sition on probability, it is enough to note that And two kinds of statistician make-up of the bag of chips. Nevertheless, uncertainty about parameters in statistical mod- for a single coin toss and a single poker chip els is almost invariably epistemic. If, for instance, On a personal note, I have been both kinds of all the uncertainty is quantifi ed in a single I was conducting experiments to measure empiri- statistician in my career. It was my experience, probability, that of Heads or red. cally the atomic weight of zinc, the unknown pa- as a young statistician, of analysing data and The difference emerges when I consider rameter is that atomic weight. I cannot consider producing frequentist tests and confi dence inter- a sequence of tosses of that coin, and a zinc as a randomly chosen element. Indeed, it vals for other scientists that convinced me that sequence of chips drawn from the bag. My is particularly zinc that I wish to know about. the Bayesian approach is the right one for sta- uncertainty about the coin tosses is still In a similar way, nearly all statistical analysis tistical analysis. I had great diffi culty persuading purely aleatory. No matter how many times is to learn about parameters, and so to reduce the scientists not to misinterpret the frequentist I toss the coin, my uncertainty about get- our epistemic uncertainty about them. Since fre- inferences I was giving them. And it was clear to ting Heads on the next toss is the same, and quentist statistics does not and cannot quantify me that this was because the correct interpreta- is expressed by a probability of 0.5. On the that uncertainty with probabilities, conventional tion was of no use to them. Frequentist infer- other hand, as I draw chips from the bag statistical inferences (such as signifi cance tests ences make only indirect statements about pa- my epistemic uncertainty about its compo- and confi dence intervals) never make probability rameters, and can only be interpreted in terms of sition reduces, and my probability for the statements about parameters. repeated sampling. Bayesian inferences directly next chip being red changes according to Yet the recipients of those conventional in- answered the scientists’ questions, making state- the chips I have now seen.
Load more

Recommended publications
  • What Is the Balance Between Certainty and Uncertainty?
    What is the balance between certainty and uncertainty? Susan Griffin, Centre for Health Economics, University of York Overview • Decision analysis – Objective function – The decision to invest/disinvest in health care programmes – The decision to require further evidence • Characterising uncertainty – Epistemic uncertainty, quality of evidence, modelling uncertainty • Possibility for research Theory Objective function • Methods can be applied to any decision problem – Objective – Quantitative index measure of outcome • Maximise health subject to budget constraint – Value improvement in quality of life in terms of equivalent improvement in length of life → QALYs • Maximise health and equity s.t budget constraint – Value improvement in equity in terms of equivalent improvement in health The decision to invest • Maximise health subject to budget constraint – Constrained optimisation problem – Given costs and health outcomes of all available programmes, identify optimal set • Marginal approach – Introducing new programme displaces existing activities – Compare health gained to health forgone resources Slope gives rate at C A which resources converted to health gains with currently funded programmes health HGA HDA resources Slope gives rate at which resources converted to health gains with currently funded programmes CB health HDB HGB Requiring more evidence • Costs and health outcomes estimated with uncertainty • Cost of uncertainty: programme funded on basis of estimated costs and outcomes might turn out not to be that which maximises health
    [Show full text]
  • Would ''Direct Realism'' Resolve the Classical Problem of Induction?
    NOU^S 38:2 (2004) 197–232 Would ‘‘Direct Realism’’ Resolve the Classical Problem of Induction? MARC LANGE University of North Carolina at Chapel Hill I Recently, there has been a modest resurgence of interest in the ‘‘Humean’’ problem of induction. For several decades following the recognized failure of Strawsonian ‘‘ordinary-language’’ dissolutions and of Wesley Salmon’s elaboration of Reichenbach’s pragmatic vindication of induction, work on the problem of induction languished. Attention turned instead toward con- firmation theory, as philosophers sensibly tried to understand precisely what it is that a justification of induction should aim to justify. Now, however, in light of Bayesian confirmation theory and other developments in epistemology, several philosophers have begun to reconsider the classical problem of induction. In section 2, I shall review a few of these developments. Though some of them will turn out to be unilluminating, others will profitably suggest that we not meet inductive scepticism by trying to justify some alleged general principle of ampliative reasoning. Accordingly, in section 3, I shall examine how the problem of induction arises in the context of one particular ‘‘inductive leap’’: the confirmation, most famously by Henrietta Leavitt and Harlow Shapley about a century ago, that a period-luminosity relation governs all Cepheid variable stars. This is a good example for the inductive sceptic’s purposes, since it is difficult to see how the sparse background knowledge available at the time could have entitled stellar astronomers to regard their observations as justifying this grand inductive generalization. I shall argue that the observation reports that confirmed the Cepheid period- luminosity law were themselves ‘‘thick’’ with expectations regarding as yet unknown laws of nature.
    [Show full text]
  • Climate Scepticism, Epistemic Dissonance, and the Ethics of Uncertainty
    SYMPOSIUM A CHANGING MORAL CLIMATE CLIMATE SCEPTICISM, EPISTEMIC DISSONANCE, AND THE ETHICS OF UNCERTAINTY BY AXEL GELFERT © 2013 – Philosophy and Public Issues (New Series), Vol. 3, No. 1 (2013): 167-208 Luiss University Press E-ISSN 2240-7987 | P-ISSN 1591-0660 [THIS PAGE INTENTIONALLY LEFT BLANK] A CHANGING MORAL CLIMATE Climate Scepticism, Epistemic Dissonance, and the Ethics of Uncertainty Axel Gelfert Abstract. When it comes to the public debate about the challenge of global climate change, moral questions are inextricably intertwined with epistemological ones. This manifests itself in at least two distinct ways. First, for a fixed set of epistemic standards, it may be irresponsible to delay policy- making until everyone agrees that such standards have been met. This has been extensively discussed in the literature on the precautionary principle. Second, key actors in the public debate may—for strategic reasons, or out of simple carelessness—engage in the selective variation of epistemic standards in response to evidence that would threaten to undermine their core beliefs, effectively leading to epistemic double standards that make rational agreement impossible. The latter scenario is aptly described as an instance of what Stephen Gardiner calls “epistemic corruption.” In the present paper, I aim to give an explanation of the cognitive basis of epistemic corruption and discuss its place within the moral landscape of the debate. In particular, I argue that epistemic corruption often reflects an agent’s attempt to reduce dissonance between incoming scientific evidence and the agent’s ideological core commitments. By selectively discounting the former, agents may attempt to preserve the coherence of the latter, yet in doing so they threaten to damage the integrity of evidence-based deliberation.
    [Show full text]
  • Investment Behaviors Under Epistemic Versus Aleatory Uncertainty
    Investment Behaviors Under Epistemic versus Aleatory Uncertainty 2 Daniel J. Walters, Assistant Professor of Marketing, INSEAD Europe, Marketing Department, Boulevard de Constance, 77300 Fontainebleau, 01 60 72 40 00, [email protected] Gülden Ülkümen, Gülden Ülkümen, Associate Professor of Marketing at University of Southern California, 701 Exposition Boulevard, Hoffman Hall 516, Los Angeles, CA 90089- 0804, [email protected] Carsten Erner, UCLA Anderson School of Management, 110 Westwood Plaza, Los Angeles, CA, USA 90095-1481, [email protected] David Tannenbaum, Assistant Professor of Management, University of Utah, Salt Lake City, UT, USA, (801) 581-8695, [email protected] Craig R. Fox, Harold Williams Professor of Management at the UCLA Anderson School of Management, Professor of Psychology and Medicine at UCLA, 110 Westwood Plaza #D511 Los Angeles, CA, USA 90095-1481, (310) 206-3403, [email protected] 3 Abstract In nine studies, we find that investors intuitively distinguish two independent dimensions of uncertainty: epistemic uncertainty that they attribute to missing knowledge, skill, or information, versus aleatory uncertainty that they attribute to chance or stochastic processes. Investors who view stock market uncertainty as higher in epistemicness (knowability) are more sensitive to their available information when choosing whether or not to invest, and are more likely to reduce uncertainty by seeking guidance from experts. In contrast, investors who view stock market uncertainty as higher
    [Show full text]
  • The Flaw of Averages)
    THE NEED FOR GREATER REFLECTION OF HETEROGENEITY IN CEA or (The fLaw of averages) Warren Stevens PhD CONFIDENTIAL © 2018 PAREXEL INTERNATIONAL CORP. Shimmering Substance Jackson Pollock, 1946. © 2018 PAREXEL INTERNATIONAL CORP. / 2 CONFIDENTIAL 1 Statistical summary of Shimmering Substance Jackson Pollock, 1946. Average color: Brown [Beige-Black] Average position of brushstroke © 2018 PAREXEL INTERNATIONAL CORP. / 3 CONFIDENTIAL WE KNOW DATA ON EFFECTIVENESS IS FAR MORE COMPLEX THAN WE LIKE TO REPRESENT IT n Many patients are better off with the comparator QALYS gained © 2018 PAREXEL INTERNATIONAL CORP. / 4 CONFIDENTIAL 2 A THOUGHT EXPERIMENT CE threshold Mean CE Mean CE treatment treatment #1 #2 n Value of Value of Cost per QALY treatment treatment #2 to Joe #1 to Bob © 2018 PAREXEL INTERNATIONAL CORP. / 5 CONFIDENTIAL A THOUGHT EXPERIMENT • Joe is receiving a treatment that is cost-effective for him, but deemed un cost-effective by/for society • Bob is receiving an treatment that is not cost-effective for him, but has been deemed cost-effective by/for society • So who’s treatment is cost-effective? © 2018 PAREXEL INTERNATIONAL CORP. / 6 CONFIDENTIAL 3 HYPOTHETICALLY THERE ARE AS MANY LEVELS OF EFFECTIVENESS (OR COST-EFFECTIVENESS) AS THERE ARE PATIENTS What questions or uncertainties stop us recognizing this basic truth in scientific practice? 1. Uncertainty (empiricism) – how do we gauge uncertainty in a sample size of 1? avoiding false positives 2. Policy – how do we institute a separate treatment policy for each and every individual? 3. Information (data) – how we do make ‘informed’ decisions without the level of detail required in terms of information (we know more about populations than we do about people) © 2018 PAREXEL INTERNATIONAL CORP.
    [Show full text]
  • The Problem of Induction
    The Problem of Induction Gilbert Harman Department of Philosophy, Princeton University Sanjeev R. Kulkarni Department of Electrical Engineering, Princeton University July 19, 2005 The Problem The problem of induction is sometimes motivated via a comparison between rules of induction and rules of deduction. Valid deductive rules are necessarily truth preserving, while inductive rules are not. So, for example, one valid deductive rule might be this: (D) From premises of the form “All F are G” and “a is F ,” the corresponding conclusion of the form “a is G” follows. The rule (D) is illustrated in the following depressing argument: (DA) All people are mortal. I am a person. So, I am mortal. The rule here is “valid” in the sense that there is no possible way in which premises satisfying the rule can be true without the corresponding conclusion also being true. A possible inductive rule might be this: (I) From premises of the form “Many many F s are known to be G,” “There are no known cases of F s that are not G,” and “a is F ,” the corresponding conclusion can be inferred of the form “a is G.” The rule (I) might be illustrated in the following “inductive argument.” (IA) Many many people are known to have been moral. There are no known cases of people who are not mortal. I am a person. So, I am mortal. 1 The rule (I) is not valid in the way that the deductive rule (D) is valid. The “premises” of the inductive inference (IA) could be true even though its “con- clusion” is not true.
    [Show full text]
  • Certainty and Uncertainty in Tax Law: Do Opposites Attract?
    laws Article Certainty and Uncertainty in Tax Law: Do Opposites Attract? Alexander V. Demin Law Institute, Siberian Federal University, 660075 Krasnoyarsk, Russia; [email protected] Received: 14 October 2020; Accepted: 26 November 2020; Published: 4 December 2020 Abstract: The principle of certainty of taxation is the dimension of a general requirement of certainty in the legal system. The purpose of this article is to argue the thesis that uncertainty in tax law is not always an absolute evil, sometimes it acts as a means of the most optimal (and in some cases the only possible) settlement of relations in the field of taxes. On the contrary, uncertainty and fragmentation in tax law are colossal problems subject to overcome by the efforts of scientists, legislators, judges, and practicing lawyers. Uncertainty in tax law is manifested in two ways: on the one hand, negatively—as a defect (omission) of the legislator and, on the other hand, positively—as a set of specific legal means and technologies that are purposefully used in lawmaking and law enforcement. In this context, relatively determined legal tools are an effective channel for transition from uncertainty to certainty in the field of taxation. A tendency towards increased use of relatively determined legal tools in lawmaking processes (for example, principles, evaluative concepts, judicial doctrines, standards of good faith and reasonableness, discretion, open-ended lists, recommendations, framework laws, silence of the law, presumptive taxation, analogy, etc.), and involving various actors (courts, law enforcement agencies and officials, international organizations, citizens, organizations and their associations) allow making tax laws more dynamic flexible, and adequate to changing realities of everyday life.
    [Show full text]
  • WP56: Scientific Uncertainty: a User Guide
    The Munich Re Programme: Evaluating the Economics of Climate Risks and Opportunities in the Insurance Sector Scientific uncertainty: a user’s guide Seamus Bradley August 2011 Centre for Climate Change Economics and Policy Working Paper No. 65 Munich Re Programme Technical Paper No. 9 Grantham Research Institute on Climate Change and the Environment Working Paper No. 56 The Centre for Climate Change Economics and Policy (CCCEP) was established by the University of Leeds and the London School of Economics and Political Science in 2008 to advance public and private action on climate change through innovative, rigorous research. The Centre is funded by the UK Economic and Social Research Council and has five inter-linked research programmes: 1. Developing climate science and economics 2. Climate change governance for a new global deal 3. Adaptation to climate change and human development 4. Governments, markets and climate change mitigation 5. The Munich Re Programme - Evaluating the economics of climate risks and opportunities in the insurance sector (funded by Munich Re) More information about the Centre for Climate Change Economics and Policy can be found at: http://www.cccep.ac.uk. The Munich Re Programme is evaluating the economics of climate risks and opportunities in the insurance sector. It is a comprehensive research programme that focuses on the assessment of the risks from climate change and on the appropriate responses, to inform decision-making in the private and public sectors. The programme is exploring, from a risk management perspective, the implications of climate change across the world, in terms of both physical impacts and regulatory responses.
    [Show full text]
  • A Philosophical Treatise of Universal Induction
    Entropy 2011, 13, 1076-1136; doi:10.3390/e13061076 OPEN ACCESS entropy ISSN 1099-4300 www.mdpi.com/journal/entropy Article A Philosophical Treatise of Universal Induction Samuel Rathmanner and Marcus Hutter ? Research School of Computer Science, Australian National University, Corner of North and Daley Road, Canberra ACT 0200, Australia ? Author to whom correspondence should be addressed; E-Mail: [email protected]. Received: 20 April 2011; in revised form: 24 May 2011 / Accepted: 27 May 2011 / Published: 3 June 2011 Abstract: Understanding inductive reasoning is a problem that has engaged mankind for thousands of years. This problem is relevant to a wide range of fields and is integral to the philosophy of science. It has been tackled by many great minds ranging from philosophers to scientists to mathematicians, and more recently computer scientists. In this article we argue the case for Solomonoff Induction, a formal inductive framework which combines algorithmic information theory with the Bayesian framework. Although it achieves excellent theoretical results and is based on solid philosophical foundations, the requisite technical knowledge necessary for understanding this framework has caused it to remain largely unknown and unappreciated in the wider scientific community. The main contribution of this article is to convey Solomonoff induction and its related concepts in a generally accessible form with the aim of bridging this current technical gap. In the process we examine the major historical contributions that have led to the formulation of Solomonoff Induction as well as criticisms of Solomonoff and induction in general. In particular we examine how Solomonoff induction addresses many issues that have plagued other inductive systems, such as the black ravens paradox and the confirmation problem, and compare this approach with other recent approaches.
    [Show full text]
  • Turns in the Evolution of the Problem of Induction*
    CARL G. HEMPEL TURNS IN THE EVOLUTION OF THE PROBLEM OF INDUCTION* 1. THE STANDARD CONCEPTION: INDUCTIVE "INFERENCE" Since the days of Hume's skeptical doubt, philosophical conceptions of the problem of induction and of ways in which it might be properly solved or dissolved have undergone a series of striking metamor- phoses. In my paper, I propose to examine some of those turnings, which seem to me to raise particularly important questions about the nature of empirical knowledge and especially scientific knowledge. Many, but by no means all, of the statements asserted by empirical science at a given time are accepted on the basis of previously established evidence sentences. Hume's skeptical doubt reflects the realization that most of those indirectly, or inferentially, accepted assertions rest on evidence that gives them no complete, no logically conclusive, support. This is, of course, the point of Hume's obser- vation that even if we have examined many occurrences of A and have found them all to be accompanied by B, it is quite conceivable, or logically possible, that some future occurrence of A might not be accompanied by B. Nor, we might add, does our evidence guarantee that past or present occurrences of A that we have not observed were- or are- accompanied by B, let alone that all occurrences ever of A are, without exception, accompanied by B. Yet, in our everyday pursuits as well as in scientific research we constantly rely on what I will call the method of inductive ac- ceptance, or MIA for short: we adopt beliefs, or expectations, about empirical matters on logically incomplete evidence, and we even base our actions on such beliefs- to the point of staking our lives on some of them.
    [Show full text]
  • Uncertainty Guide
    ICSBEP Guide to the Expression of Uncertainties Editor V. F. Dean Subcontractor to Idaho National Laboratory Independent Reviewer Larry G. Blackwood (retired) Idaho National Laboratory Guide to the Expression of Uncertainties for the Evaluation of Critical Experiments ACKNOWLEDGMENT We are most grateful to Fritz H. Fröhner, Kernforschungszentrum Karlsruhe, Institut für Neutronenphysik und Reaktortechnik, for his preliminary review of this document and for his helpful comments and answers to questions. Revision: 4 i Date: November 30, 2007 Guide to the Expression of Uncertainties for the Evaluation of Critical Experiments This guide was prepared by an ICSBEP subgroup led by G. Poullot, D. Doutriaux, and J. Anno of IRSN, France. The subgroup membership includes the following individuals: J. Anno IRSN (retired), France J. B. Briggs INL, USA R. Bartholomay Westinghouse Safety Management Solutions, USA V. F. Dean, editor Subcontractor to INL, USA D. Doutriaux IRSN (retired), France K. Elam ORNL, USA C. Hopper ORNL, USA R. Jeraj J. Stefan Institute, Slovenia Y. Kravchenko RRC Kurchatov Institute, Russian Federation V. Lyutov VNIITF, Russian Federation R. D. McKnight ANL, USA Y. Miyoshi JAERI, Japan R. D. Mosteller LANL, USA A. Nouri OECD Nuclear Energy Agency, France G. Poullot IRSN (retired), France R. W. Schaefer ANL (retired), USA N. Smith AEAT, United Kingdom Z. Szatmary Technical University of Budapest, Hungary F. Trumble Westinghouse Safety Management Solutions, USA A. Tsiboulia IPPE, Russian Federation O. Zurron ENUSA, Spain The contribution of French participants had strong technical support from Benoit Queffelec, Director of the Société Industrielle d’Etudes et Réalisations, Enghien (France). He is a consultant on statistics applied to industrial and fabrication processes.
    [Show full text]
  • MARLAP Manual Volume III: Chapter 19, Measurement Uncertainty
    19 MEASUREMENT UNCERTAINTY 19.1 Overview This chapter discusses the evaluation and reporting of measurement uncertainty. Laboratory measurements always involve uncertainty, which must be considered when analytical results are used as part of a basis for making decisions.1 Every measured result reported by a laboratory should be accompanied by an explicit uncertainty estimate. One purpose of this chapter is to give users of radioanalytical data an understanding of the causes of measurement uncertainty and of the meaning of uncertainty statements in laboratory reports. The chapter also describes proce- dures which laboratory personnel use to estimate uncertainties. This chapter has more than one intended audience. Not all readers are expected to have the mathematical skills necessary to read and completely understand the entire chapter. For this reason the material is arranged so that general information is presented first and the more tech- nical information, which is intended primarily for laboratory personnel with the required mathe- matical skills, is presented last. The general discussion in Sections 19.2 and 19.3 requires little previous knowledge of statistical metrology on the part of the reader and involves no mathe- matical formulas; however, if the reader is unfamiliar with the fundamental concepts and terms of probability and statistics, he or she should read Attachment 19A before starting Section 19.3. The technical discussion in Sections 19.4 and 19.5 requires an understanding of basic algebra and at least some familiarity with the fundamental concepts of probability and statistics. The discus- sion of uncertainty propagation requires knowledge of differential calculus for a com- Contents plete understanding.
    [Show full text]