Bibliography
§5.1, p. 93, doesn’t know which axiom to add to additive representability to get SEU. Well, that’s exactly the tradeoff consistency that I wrote in several papers!
Download 7.23 Mb.
|
- Navigate this page:
- Newcomb ’ s paradox
| §5.1, p. 93, doesn’t know which axiom to add to additive representability to get SEU. Well, that’s exactly the tradeoff consistency that I wrote in several papers!
My personal opinions agree very well with those in this book. I don’t know any other work in the literature that agrees so much with my subjective viewpoints: - That separability and EU are normative for uncertainty, to a lesser degree for interpersonal and intertemporal (intertemporal separability criticized), - that completeness of preference is the weakest of the axioms, - that there should be one cardinal utility index for all social science (“risky = riskless utility”) although there exists no conclusive argument for or against that yet, - that the “tradeoff thinking” helps support the above viewpoint of one cardinal utility for all social sciences. Reviews of Broome (1991): %{ positive. Doesn’t want moral aspects included. Suggests that equity is part of the goodness measure but not of utility. %} Cowen, Tyler (1992) “Weighing Goods: Equality, Uncertainty, and Time,” Book Review of: John R. Broome (1991) “Weighing Goods,” Basil Blackwell, Oxford, UK; Economics and Philosophy 8, 283–284. %{ Nice summary of Broome’s arguments on separability, goodness, completeness, etc., with some own opinions added. %} Hausman, Daniel M. (1993) “The Structure of Good,” Ethics 103, 792–806. %{ (positive and presents main themes; doesn’t try to be deep.) %} Hollis, Martin (1992) Mind 101, 553–554. Sugden, Robert (1992) Economica 59, 253–254. Pattanaik, Prasanta K. (1993) Economic Journal 103, 752–753. Arneson, Richard J. (1993) Journal of Economic Literature 31, 1443–1445. This review is superficial; in particular the listing of arguments against utilitarianism, at the end, is out of place as Broome’s book discusses each of them extensively. %} Temkin, Larry S. (1994) Philosophy and Public Affairs 23, 350–380. %} Broome, John R. (1991) “Weighing Goods.” Basil Blackwell, Oxford, UK.
Broome, John R. (1991) “Desire, Belief and Expectation,” Mind 100, 265–267. {% %} Broome, John R. (1991) “Utility,” Economics and Philosophy 7, 1–12. {% %} Broome, John R. (1991) “A Reply to Sen,” Economics and Philosophy 7, 285–287. {% paternalism/Humean-view-of-preference: Humean viewpoint: no preference can ever be criticized for being irrational. Moderate Humean viewpoint: only internal consistency conditions (such as transitivity) can be imposed, no other criteria for rationality. Broome argues that the moderate Humean viewpoint cannot be maintained in the sense that it must necessarily reduce to the Humean viewpoint, as follows. Violations of internal consistency can always be avoided by remodeling, by “finer individuation” of alternatives (e.g., incorporating context-dependence in the description of the alternative). Such finer individuation cannot be criticized on the basis of internal consistency and must necessarily be discussed on external grounds. I personally think that both the Humean and the moderate Humean viewpoint are untenable, and that external criteria have to be invoked in rationality. The viewpoint that only the consistency axioms, and not for instance medical knowledge, is required for rationality, is surely not fruitful in medical decision making! Probably Broome thinks the same, see end of §1: “I hope this will diminish the appeal of the Humean view as a whole.” P. 58 (on a book-making reasoning): “It is as though you stole his shirt and then sold it back to him.” P. 65: “a person’s practical preferences are causally affected by her nonpractical preferences” %} Broome, John R. (1993) “Can a Humean Be Moderate?” In Raymond G. Frey & Christopher W. Morris (eds.) Value, Welfare and Morality. Cambridge University Press, Cambridge. {% An abbreviated description of the ideas of Broome (1991), with implications for QALYs. P. 150 2nd para: claims that EU is normative intertemporal separability criticized: pp. 151–152 Pp. 153-154: risky utility u = transform of strength of preference v; states this point not for strength of preference but for intertemporal utility used in discounted utility. Pp. 154-155: risky utility u = transform of strength of preference v; states this point not for strength of preference but for intertemporal utility versus a general cardinal index of utility, called “good” by the author. Pp. 155 bottom: risky utility u = transform of strength of preference v; states this point not for strength of preference but for a general cardinal index of utility, called “good” by the author, versus EU utility. P. 156 bottom suggests that intertemporal utility has more right to claim to be a cardinal index of goodness than risky EU utility. No argument is given, but the opinion is repeated three times or so. P. 154 3rd para distinguishes cardinal in the mathematical sense from cardinal in the sense of index of goodness. questionnaire versus choice utility: pp. 159-160 suggest that direct judgment may be better for measuring a normative index of goodness that eliciting preferences. %} Broome, John R. (1993) “Qalys,” Journal of Public Economics 50, 149–167. {% %} Broome, John R. (1993) “Goodness Is Reducible to Betterness: The Evil of Death Is the Value of Life,” - Discussion Papers (University of Bristol, Department of Economics) {% %} Broome, John (1999) “Ethics out of Economics.” Cambridge University Press, Cambridge. {% risky utility u = strength of preference v (or other riskless cardinal utility, often called value): Broome believes so at the level of his degree of goodness (§§6 and 7); calls it the expectational concept. His argument is that this is most natural and that there is no natural alternative. He seems not to believe so at the level of rightness (§2), where he says that risk neutrality for rightness in goodness is not plausible. %} Broome, John R. (2008) “Can There Be a Preference-Based Utilitarianism.” In Maurice Salles & John Weymark (eds.) Justice, Political Liberalism and Utilitarianism: Themes from Harsanyi and Rawls, 221–238, Cambridge University Press, Cambridge. {% discounting normative; extensively discuss whether or not we ought to discount. Have no strong position, but favor discounting. %} Broome, John R. & David Ulph (1992) “Counting the Cost of Global Warming: A Report to the Economic and Social Research Council.” White University Press, Cambridge. {% Propose a measure for how much information about unknown subjective parameters to be measured a set of decision problems gives. Such measures are used in recent computer-based adaptive measurements where the new stimulus offered to the subject is chosen to give optimal info given previous choices of the subject, as for instance in Cavagnaro, Gonzalez, Myung, & Pitt (2013, Management Science). But now the criterion is simpler and more tractable, and does not depend on previous choices. I expect that TO method based measurements do well. They apply their method to the measurement of PT (they write CPT; I mean the 1992 version of their theory). They then use power utility and the 1-parameter Prelec probability weighting family. Pp. 265 ff. show that variations/errors in observations contribute to the DFD-DFE gap, due to positive skewedness and the lower bound of 0. They find, surprisingly, that the stimulus set deliberately chosen by Stott (2006) in fact has more overlap of data estimation than a randomly constructed set by Erev et al. (2002) (p 268 bottom). Unfortunately, this paper follows the bad terminology of some papers in DFE to let “diminishing sensitivity” refer only to utility curvature and even equate the two. This paper finds, again, that estimations of loss aversion are not stable. The authors add an argument to the many existing: that there often are not many mixed lotteries and only those contribute to the estimation (p. 269 3rd para). %} Broomell, Stephen B. & Sudeep Bhatia (2015) “Parameter Recovery for Decision Modeling Using Choice Data,” Decision 1, 252–274. {% Seem to find that the strategy method gives different results than posterior choice. %} Brosig, Jeanette., Joachim Weimann, & Chun-Lei Yang (2003) “The Hot versus Cold Effect in a Simple Bargaining Experiment,” Experimental Economics 6, 75–90. {% They find endowment effect with 20 chimpanzees for objects of value. Unsurprisingly, there is no discrepancy for objects that are of no value anyhow. %} Brosnan, Sarah F., Owen D. Jones, Molly Gardner, Susan P. Lambeth, & Steven J. Schapiro (2012) “Evolution and the Expression of Biases: Situational Value Changes the Endowment Effect in Chimpanzees,” Evolution and Human Behavior 33, 378–386. {% %} Brothers, Alan (1990) “An Empirical Investigation of Some Properties that are Relevant to Generalized Expected Utility Theory,” doctoral dissertation, University of California, Irvine. {% Replicate Plott & Zeiler but without anonymity, showing that familiarity with the procedures drives it rather than anonymity. %} Brown, Alexander L. & Gregory Cohen (2015) “Does Anonymity Affect the Willingness to Accept and Willingness to Pay Gap? A Generalization of Plott and Zeiler,” Experimental Economics 18, 173–184. {% Test Epstein-Zin preferences. %} Brown, Alexander L. & Hwagyun Kim (2014) “Do Individuals Have Preferences Used in Macro-Finance Models? An Experimental Investigation,” Management Science 60, 939–958. {% Consider DUU with real outcomes so outcome-wise mixing. Consider risk measures, being functionals that satisfy translation invariance (= constant absolute risk aversion = homotheticity), convexity, and some other properties, and discuss many examples satisfying these conditions such as CEU (Choquet expected utility) with proper restrictions. %} Brown, David B., Enrico De Giorgi, & Melvyn Sim (2012) “Aspirational Preferences and Their Representation by Risk Measures,” Management Science 58, 2095–2113. {% A generalization of more risk averse of Rotschild-Stiglitz, allowing for a sort of positive correlation between the noise-lottery added and the base lottery. %} Brown, David P. (2017) New Characterizations of Increasing Risk,” Journal of Mathematical Economics 69, 7–11. {% Find evidence for rank dependence %} Brown, Gordon D.A., Jonathan Gardner, Andrew J. Oswald, & Jing Qian (2008) “Does Wage Rank Affect Employees’ Well-Being?,” Industrial Relations 47, 355–389. {% Version of April ’04: inverse-S: beginning has nice survey. Paper discusses large and small probabilities without relating them to outcomes/rank-dependence. Use range-frequency theory (RFT) of Parducci (1965, 1995) to explain inverse-S probability weighing. According to RFT, we are extra sensitive to stimuli in regions where there are many observations/experiences, and insensitive in regions where there are few. Thus, if we more often encounter small and large probabilities, then we will be extra sensitive towards them. Difficulty is, what should we take as set of experiences? All probabilities we ever saw in our life, all probabilities occurring in the experiment we participate in so far, or only the probabilities occurring in the prospect now considered. The theoretical discussion is nice, but testing these hypotheses empirically is not easy. The authors nevertheless try and, e.g., answer how frequent they think that probabilities appear. %} Brown, Gordon D.A. & Jing Qian (2004) “The Origin of Probability Weighting: A Psychophysical Approach,” University of Warwick. {% P. 489 seems to argue for conditioning on ancillary statistic, and cite Fisher, Savage, Cox %} Brown, Lawrence D. (1990) “An Ancillarity Paradox Which Appears in Multiple Linear Regression” (including discussion), Annals of Statistics 18, 471–538. {% foundations of quantum mechanics %} Brown, Matthew J. (2009) “Relational Quantum Mechanics and the Determinacy Problem,” The British Journal for the Philosophy of Science 60, 679–695. {% conditional probability %} Brown, Peter M. (1976) “Conditionalization and Expected Utility,” Philosophy of Science 43, 415–419. {% %} Brown, Roger (1965) “Social Psychology.” New York: Free Press. {% There were 5 hypothetical risky decision questions (imagine your income would either double or …), used to measure risk attitudes. They are negatively related to their children’s test scores and attending college post high scholes. %} Brown, Sarah, Aurora Ortiz- Nuñez & Karl Taylor (2012) “Parental Risk Attitudes and Children’s Academic Test Scores: Evidence from the US Panel Study of Income Dynamics,” Scottish Journal of Political Economy 59, 47–70. {% Z&Z %} Browne, Mark J. & Helene I. Doerpinghaus (1993) “Information Asymmetries and Adverse Selection in the Market for Individual Medical Expense Insurance,” Journal of Risk and Insurance 60, 300–312. {% Use data from a portfolio of risks of a German insurer. Within-subject comparions give that people rather insure their bike than their house against floods. %} Browne, Mark J., Christian Knoller, & Andreas Richter (2015) “Behavioral Bias and the Demand for Bicycle and Flood Insurance,” Journal of Risk and Uncertainty 50, 141–160. {% foundations of statistics %} Browner, Warren S. & Thomas B. Newman (1987) “Are All Significant P values Created Equal? The Analogy between Diagnostic Tests and Clinical Research,” Journal of the American Medical Association 257, 2459–2463. {% intertemporal separability criticized: %} Browning, Martin (1991) “A Simple Nonadditive Preference Structure for Models of Household Behavior over Time,” Journal of Political Economy 99, 6707–637. {% Analyze data from the Canadian Family Expenditure Survey in 7 years between 1974 and 1992, assuming that households cannot be considered as one, but are composed of different individuals. Nicely, Slutsky symmetry, a necessary condition for utility maximization, is not rejected for singles (p. 1245), but is for general families. %} Browning, Martin & Pierre Andre Chiappori (1998) “Efficients Intra-Household Allocations: A General Characterization and Empirical Tests,” Econometrica 66, 1241–1278. {% %} Brouwer, Luitzen E.J. (1911) “Über Abbildung von Mannigfaltigkeiten,” Mathematische Annalen 71, 97–115. {% %} Brouwer, Luitzen E.J. (1924) “Beweis das Jede Volle Funktion Gleichmässig Stetig Ist,” Proceedings KNAW 27, 189–194. Reprinted in Arend Heyting, (1975, ed.) “Collected Works,” Vol. I, 478–479. North-Holland, Amsterdam. {% Seems to add refinements to Bruhin, Fehr-Duda, & Epper (2007), using the same data set. The abstract and intro mostly repeat the general points of the preceding paper and I did not see what was new here. %} Bruhin, Adrian, (2008) “Stochastic Epected Utility and Prospect Theory in a Horse Race: A Finite Mixture Approach,” working paper. {% Uses real incentives for gains; losses from prior endowment mechanism; Zurich 2003 179 subjects, 50 lotteries Zurich 2006 118 subjects, 40 lotteries Bejing Nov. 2005 151 subjects, 28 lotteries Determine CEs (certainty equivalents) from choice lists, and fit PT. Do mixture models. Optimal result is with 2 groups, one (20%) doing EV and the other doing PT with all the patterns of T&K 92 confirmed: concave utility for gains, convex utility for losses; inverse-S; find it using Goldstein & Einhorn (1987) family. Risk averse for gains, risk seeking for losses reflection at individual level for risk: it is in their data but they do not report it. Have no mixed prospects and, hence, model and measure no loss aversion. For gains, Chinese students are less pessimistic and more likelihood insensitive than Swiss students. They also have more concave utility and, because CE data may not separate utility well from probability weighting (collinearity), it was not clear to me to what extent the higher concavity of utility drives the lower probability weighting. The authors are happy about each subject clearly falling into one of the two categories (w, probability weighting, linear or nonlinear). I did not understand what else could happen than these two. There are few subjects of the “ambiguous type” (between the two categories, with p = 0.4 of being one catefory and p = 0.6 of being the other, as an example they give) but I don’t know if their probabilistic models give much space to such types in, say, randomly generated choices for instance. %} Bruhin, Adrian, Helga Fehr-Duda, & Thomas Epper (2010) “Risk and Rationality: Uncovering Heterogeneity in Probability Distortion,” Econometrica 78, 1375–1412. {% inverse-S: fifty-fifty is Principle of Complete Ignorance is extreme case of inverse-S. This paper conjectures, and finds confirmed, that more fifty-fifty reasoning occurs (a) for singular than for distributional formats (b) less controlable events (c) less numerate respondents (d) less educated respondents. (cognitive ability related to likelihood insensitivity (= inverse-S)) (c) remains after correction for age and education. %} Bruine de Bruin, Wändi, Baruch Fischhoff, Susan G. Millstein, & Bonnie L. Halpern-Felsher (2000) “Verbal and Numerical Expressions of Probability: 'It's a Fifty-Fifty Chance',” Organizational Behavior and Human Decision Processes 81, 115–131. {% Subjects can estimate probabilities in percentages. Those that estimate 0% get a refined scale for probabilities close to 0 and, obviously, many then go some above 0. %} Bruine de Bruin, Wändi, Andrew M. Parker, & Jürgen Maurer (2011) “Assessing Small non-Zero Perceptions of Chance: The Case of H1N1 (Swine) Flu Risks,” Journal of Risk and Uncertainty 42, 145–159. {% (Algemeen Dagblad is a daily newspaper, with 300,000 copies per day, and is the 2nd largest newspaper in the Netherlands.) %} Bruinsma, Gea & Peter P. Wakker (2017) “Ook naar het Strand Neem Ik Werk mee,” Algemeen Dagblad 8 August 2017, Beurs 17. {% People prefer to predict unknown result of toss of coin before toss to after toss. So source dependence of information relates to timing, although it here always is known probability. It, hence, provides a case where known probability is not really one source. Introduction gives references to source-preferences. This paper argues that the difference between pre- and post-diction, usually ascribed to magical thinking, can have other causes, using open-ended questions to subjects to find out. The authors find many other causes, but point out a limitation to their study on p. 24 l. 3: “Of course, we cannot rule out the possibility that some subjects might have been reluctant to disclose their belief in magic.” %} Brun, Wibecke & Karl H. Teigen (1990) “Prediction and Postdiction Preferences in Guessing,” Journal of Behavioral Decision Making 3, 17–28. {% If expected value can be increased by increasing probability or increasing outcome, then what will subjects prefer? The author tests it. %} Bruner, David M. (2009) “Changing the Probability versus Changing the Reward,” Experimental Economics 12, 367–385. {% Shows that decision error decreases with risk aversion. %} Bruner, David M. (2017) “Does Decision Error Decrease with Risk Aversion?,” Experimental Economics 259–273. {% %} Bruner, Jerome S. & Cecile C. Goodman (1947) “Value and Need as Organizing Factors in Perception,” Journal of Abnormal and Social Psychology 42, 33–44. {% DOI: http://dx.doi.org/10.1007/s11238-014-9432-5 Study risk aversion (measured through choice list of Holt & Laury 2002), and ambiguity aversion, choosing from known/unknown urn. Do it individually, group process of unanimity rule, and group process of majority. Find increased risk aversion in group processes, but no significant differences for ambiguity attitude. The authors use the smooth model to analyze ambiguity through parameter s in Table 3, but I did not see specified how they chose the second-order probabilities. %} Brunette, Marielle, Laure Cabantous, & Stéphane Couture (2015) “Are Individuals More Risk and Ambiguity Averse in a Group Environment or Alone? Results from an Experimental Study,” Theory and Decision 78, 357–376. {% Use smooth model to analyze that, for instance, ambiguity aversion increases demand of insurance. They test particular theoretical inequalities in an experiment. %} Brunette, Marielle, Laure Cabantous, Stéphane Couture, & Anne Stenger (2013) “The Impact of Governmental Assistance on Insurance Demand under Ambiguity: A Theoretical Model and an Experimental Test,” Theory and Decision 75, 153–174. {% questionnaire versus choice utility %} Bruni, Luigino & Francesco Guala 2001) “Pareto and the Epistemological Foundations of Rational Choice,” History of Political Economy 33, 21–49. {% questionnaire versus choice utility The authors discuss Pareto’s views on utility, and connect them to modern issues, in particular Plott’s discovered preference hypothesis. To cite someone opposed to Pareto, they often cite Pantaleoni. On a few points I disagree with the authors: 1. They assume that behavioral economists do not accept the revealed-preference paradigm but want introspective psychological inputs. The same claim is made by Angner & Loewenstein (2010). I think that the link is less strong, and disagree with both these teams. Behavioral economists point out problems for revealed preference, are often close to psychologists, and their work gives support to abandoning revealed preference. But behavioral economics does not necessarily abandon revealed preference. It is still essentially within the revealed preference paradigm, showing there are more problems there than thought but yet to be resolved. For example, virtually all papers by Kahneman & Tversky use only revealed preference inputs. 2. I disagree much with the suggestion, on p. 152 ff., that part of diminishing sensitivity correspond to reference dependence. State-dependent reference points is a research interest of Sugden (e.g. his 2003-JET paper), but he/they got carried away thinking that Edgeworth's diminishing marginal utility be that. On p. 153 the authors write: “it is surely significant that he [Edgeworth] was aware of the reference-dependence of preferences, ..” It concerns the point that if I ate 2 apples each of the last 10 days, then I like an apple less today than if I didn't eat any for 10 days, an aspect of diminishing marginal utility put forward by Edworth. Contrary to the suggestions of Bruni & Sugden, this is not reference dependence. It is simply intertemporal dependence, dependence on PHYSICAL CIRCUMSTANCES. It is completely standard in economic analyses. Reference-dependence concerns only framing situations, where the physical circumstances are the same but the PSYCHOLOGICAL PERCEPTION is different, something which is not standard in economic analyses. §6 criticizes the discovered preference hypothesis, arguing that (1) if preference converge after learning the limit need not be true preference but may be ad-hoc learned heuristic (the shaping hypothesis); (2) many choices in our life must be made without chance to learn from repetition; (3.a) even if people learn preferences, these need not be consistent or context independent; (3.b) in substantive justification of consistency, amounting to assumption that people maximize some (objectively measurable) index such as happiness, how justify this measure? Probably requires resort to psychology, exactly the thing that Pareto and many economists don’t want. %} Bruni, Luigino & Robert Sugden (2007) “The Road not Taken: How Psychology Was Removed from Economics, and how It Might Be Brought Back,” Economic Journal 117, 146–175. {% If consumer is not certain to find optimal consumption bundle, then this can generate risk aversion for gains but risk seeking for losses, as posited by prospect theory. %} Brunnermeier, Markus K. (2004) “Learning to Reoptimize Consumption at New Income Levels: A Rationale for Prospect Theory,” Journal of the European Economic Association 2, 98–114. {% decreasing ARA/increasing RRA: find support for constant RRA (p. 714 4th para; p. 734 ) + inertia (p. 714 last para; p. 734), and against habit formation (p. 733 §III 1st para);. Use household-level panel data from the Panel Study of Income Dynamics, covering a period of about 20 years (p. 714). %} Brunnermeier, Markus K. & Stefan Nagel (2008) “Do Wealth Fluctuations Generate Time-Varying Risk Aversion? Micro-Evidence on Individuals’ Asset Allocation,” American Economic Review 98, 713–736. {% In this paper, subjective probabilities (beliefs) can be chosen so as to maximize utility. For instance, in a prospect 1000.50 you can believe that you get 100 with probability 1 and thus get the highest possible (expected) utility, so this is what you then do. It is a Baron von Münchhausen way to get more utility. (He got himself out of a hole by very strongly, with his own hand, pulling his shoe leashes, thus lifting himself up, at least this is how his own story goes.) However, if decisions are to be taken then such misbeliefs can lead to suboptimal decisions. Then the optimal tradeoff between decision utility lost, and Baron-von-Münchhausen utility gained, has to be made. %} Brunnermeier, Markus K. & Jonathan A. Parker (2005) “Optimal Expectations,” American Economic Review 95, 1092–1118. {% %} Bruno, James E. & Arie Dirkzwager (1995) “Determining the Optimal Number of Alternatives to a Multiple-Choice Test Item: An Information Theoretic Perspective,” Educational and Psychological Measurement 55, 959–966. {% Does this paper contain the famous model? %} Brunswik, Egon (1952) “The Conceptual Framework of Psychology.” In International Encyclopedia of Unified Science, 1 (10), University Press of Chicago, Chicago. {% real incentives/hypothetical choice: compare them in the health domain and find no difference, supporting the use of hypothetical choice. NB = 179 patients were asked hypothetical WTP for self-management equipment for testing blood for anticoagulation therapy. They did not know that later they got the change to really buy. The actual decisions were well consistent with the hypothetical declarations. %} Bryan, Stirling & Sue Jowett (2010) “Hypothetical versus Real Preferences: Results from an Opportunistic Field Experiment,” Health Economics 19, 1502–1509. {% Updated for new releases of SPSS %} Bryman, Alan & Duncan Cramer (1999) “Quantitative Data Analysis with SPSS Release 8 for Windows.” Routledge, London. {% %} Brysbaert, Marc, Wim Fias, & Marie-Pascale Noël (1998) “The Whorfian Hypothesis and Numerical Cognition: Is “Twenty-Four” Processed in the Same Way as “Four-and-Twenty”?,” Cognition 66, 51–77. {% Nudge shows that in some situations behavioral economics (BE) can lead to improvements of decisions with no, or very minimal, paternalism. This is remarkable because it proves that behavioral economics can have some things to offer without commitment to paternalism. It obviously does not say that BE should do this in all situations, or that in all situations paternalism should be avoided. In many situations it can’t. Li, Li, & Wakker (2014, Theory and Decision) argue for this point. The authors here discuss behavioral law economics (BLE), and seem to equate it with nudge. Then they go at great length to argue for the obvious: that nudge does not work in all situations, and that paternalism and optimization beyond nudge shouldn’t always be avoided. %} Bubb, Ryan & Richard H. Pildes (2014) “How Behavioral Economics Trims Its Sails and Why,” Harvard Law Review 127, 1593–1678. {% information aversion: paper assumes RDU with probabilistic sophistication as normative, as in her other works, but points out that the argument holds in general. She then shows how nonEU can lead to aversion to info, and gives philosophical background. It would be nice if she would explicitly relate to the dynamic dec ision principles of Machina (1989 JEL). %} Buchak, Lara (2012) “Instrumental Rationality, Epistemic Rationality, and Evidence-Gathering,” Philosophical Perspectives 24, 85–120. {% She defines faith as accepting something and not being willing to/not being nterested in search for falsifying evidence. She justifies the latter by her work (2012 Philosophical Perspectives) on aversion to info which can happen under nonEU. (information aversion) %} Buchak, Lara (2012) “Can It Be Rational to Have Faith?.” In Jake Chandler & Victoria S. Harrison (eds.), Probability in the Philosophy of Religion, 225–247, Oxford University Press, New York. {% Tradeoff method: is used in axiomatizations. Axiomatizes probabilistically sophisticated RDU under uncertainty; i.e., Quiggin’s RDU for risk only now with the probabilities subjective, derived from acts. The author argues for this as a rational model. Many philosophic discussions on interpretations, normative status, and so on. %} Buchak, Lara (2013) “Risk and Rationality.” Oxford University Press, New York. {% free-will/determinism: takes issue with Van Inwagen’s rollback argument (see my comments at his paper). Argues that indeterminism can lead to free will in ways different than probability/chance. %} Buchak, Lara (2013) “Free Acts and Chance: Why the Rollback Argument Fails,” Philosophical Quarterly 63, 20–28. {% DOI: http://dx.doi.org/10.1007/s11098-013-0182-y On blaming and the necessity or not to use information beyond doubt (credence) or partial beliefs there. %} Buchak, Lara (2014) “Belief, Credence, and Norms,” Philosophical Studies 169, 285–311. {% Tradeoff method: is used in axiomatizations. This paper discusses the author’s preferred REU model, and its axioms. %} Buchak, Lara (2014) “Risk and Tradeoffs,” Erkenntnis 79, 1091–1117. {% Newcomb’s paradox: there is quite a bit of this, with nuances on different kinds of causality and causal decision theory. Discusses preference axiomatizations, their normative and descriptive status, but also their interpretive status. The latter means that we interpret, for instance, subjective probabilities and utilities derived from decisions as reflecting the state of the decision maker, and as genuine beliefs and happiness. If deviation from EU, the descriptive approach will simply turn to other model. The interpretive view will not do so, because beliefs and happiness are taken to be as in EU (almost by definition). They will rather search for alternative interpretations such as taking outcomes more complex. The interpretive view says that preferences deviating from EU (or whatever is taken as the appropriate theory) do not really reflect the preferences of the agent. They search for an idealized version of the agent. I am sympathetic to this view. %} Buchak, Lara (2016) “Decision Theory.” In Christopher Hitchcock & Alan Hájek (2016, eds.) Oxford Handbook of Probability and Philosophy, @–@, Oxford University Press, New York, forthcoming. {% P. 116 (citation from Sen): “Rationality or irrationality as an attribute of the social group implies the imputation to that group of an organic existence apart from that of its individual components” %} Buchanan, James M. (1954) “Social Choice, Democracy and Free Markets,” Journal of Political Economy 62, 114–123. {% utility elicitation %} Buckingham, Kenneth J. (1993) “Risks in Utility Assessment and Risks of Medical Interventions,” Medical Decision Making 13, 167–168. {% utility elicitation %} Buckingham, Kenneth J. (1993) “A Note on HYE (Healthy Years Equivalent),” Journal of Health Economics 11, 301–309. {% E0 ~ p0, for > 0, defines objective probability p as the matching probability of event E. If a person does not do EU but weights probabilities, and does so the same way for objective and subjective probabilities, then the matching probability p still is the subjective probability of E. (P.s.: even, more generally, under all probabilistic sophistication.) However, if the weighting function is different for objective probabilities than for subjective ones (as in the source method of Abdellaoui et al. 2011 AER), then this is not so. This is what this paper points out. It calculates through many examples with many weighting functions to illustrate this point many times. This is what this paper does. %} Budescu, David, Ali Abbas, & Lijuan Wu (2011) “Does Probability Weighting Matter in Probability Elicitation?,” Journal of Mathematical Psychology 55, 320–327. {% proper scoring rules: not really that, and rather scoring of exams in education, but with many related debates. For example, that even if two scoring rules are equivalent and only linear transformations of each other, one that uses loss scores may be perceived differently (p. 285). And points like if there is a critical level to pass, subjects may have to be risk seeking or risk averse (p. 283 ff.). And that it may be a burden to the subjects just to understand the strategic aspects of the scoring rule, and to be aware of their level of knowledge (p. 278 penultimate para and elsewhere). %} Budescu, David V. & Maya Bar-Hillel (1993) “To Guess or not to Guess: A Decision-Theoretic View of Formula Scoring,” Journal of Educational Management 4, 277–291. {% It is well-known that in expert aggregation, it is sometimes better to combine the best and, say, the 3rd best expert, rather than the best and the 2nd best expert, because the latter two are too closely related to each other and don’t add much to each other. This is the starting point of this paper. It proposes to select experts on the basis of how much their marginal contribution is to the rest of the group. Contribution can be measured, for instance, in terms of a proper scoring rule applied to some aggregation of the experts. The paper presents three data sets where their measure performs better than taking the best experts based on past performance. Topic for future research is to find out how general this superiority is or to what extent it was just because of the data sets chosen. May be some theoretical observations on when this approach is better than others and when not. Note that instead of marginal individual contribution, many other contribution indexses could be considered. Cooperative game theory has many proposals, such as the Shapley value. %} Budescu, David V. & Eva Chen (2015) “Identifying Expertise to Extract the Wisdom of Crowds,” Management Science 61, 267–280. {% dynamic consistency: test of RCLA %} Budescu, David V. & Ilan Fischer (2001) “The Same but Different: An Empirical Investigation of the Reducibility Principle,” Journal of Behavioral Decision Making 14, 187–206. {% Take lotteries with vague probabilities (“probability is between 0.03 and .07”), or with vague outcomes (“gain is between $45 and $105”; ambiguous outcomes vs. ambiguous probabilities). Common decision theories could take this as two-stage uncertainty, where the second stage is nonprobabilized. For vague outcomes, the authors evaluate the second stage not by w1U(x1) + (1w1)U(x2) etc. as common theories would do it, but by U(w1x1 + (1w1)x2). Could be interpreted as a very special case of Kreps & Porteus (1978). For vague probabilities they do a similar w'1p1 + (1w'1)p2, where the w1 and w1' are indexes of optimism/pessimism. Could be rephrased as rank-dependent probability transformation. They ask for certainty equivalents. Probably because of scale compatibility, as the authors mention on some occasions but not on others, the subjects are thereby more sensitive towards vagueness in outcomes. Download 7.23 Mb. Share with your friends:
|