Bibliography
Download 7.23 Mb.
|
- Navigate this page:
- Total utility theory ; questionnaire versus choice utility
- Principle of Complete Ignorance
- Risk averse for gains, risk seeking for losses
Dutch book: Ch. 3 is on de Finetti’s book making argument. Elementary introduction into decision theory, emphasizing conceptual logical and philosophical issues. Reviewed in Philosophical Review XCIX, 1990, 272–275. %} Resnik, Michael (1987) “Choices: An Introduction to Decision Theory.” University of Minnesota Press, Minneapolis, MN. {% measure of similarity %} Resnik, Philip (1999) “Semantic Similarity in a Taxonomy: An Information-Based Measure and Its Application to Problems of Ambiguity and Natural Language,” Journal of Artificial Intelligence Research 11, 95–130. {% Measuring subjective discounting for money has the problem that money is fungible: can be saved in the bank at market interest rate. So this paper compares it with subjective discounting for chocolate and so on, being things that are not fungible. It finds significant correlations, which give some support for money being usable for measuring subjective discounting. %} Reuben, Ernesto, Paolo Sapienza, & Luigi Zingales (2010) “Time Discounting for Primary and Monetary Rewards,” Economic Letters 106, 125–127. {% %} Reutskaja, Elena, Rosemarie Nagel, Colin F. Camerer, & Antonio Rangel (2011) “Search Dynamics in Consumer Choice under Time Pressure: An Eye-Tracking Study,” American Economic Review 101, 900–926. {% %} Reve, Gerard Cornelis van het (1967) “Veertien Etsen van Frans Lodewijk Pannekoek voor Arbeiders Verklaard.” Rapenburg, Amsterdam. {% Total utility theory; questionnaire versus choice utility: in this review, 15 studies are mentioned that have done both utility measurement and psychometric measurement; TTO typically has R2 of .18 till .43 with valuations of health status scales. SG doesn’t do well: SG is worse, .07 to .30. Note that we should not expect overly high correlations because of interindividual variation in the use of response scales. %} Revicki, Dennis A. & Robert M. Kaplan (1993) “Relationship between Psychometric and Utility-Based Approaches to the Measurement of Health-Related Quality of Life,” Quality Life Research 2, 477–487. {% %} Revuz, André (1955-56) “Fonctions Croissantes et Mesures sur les Espaces Topologiques Ordonnés,” Anneles de l’Institut Fourier 6, 187–268. {% Principle of Complete Ignorance: p. 11 inverse-S: this paper discusses in much detail the psychology of being more or less sensitive to numerical scales, and the ability to more or less discriminate between options, and maybe taking numbers only as categories. I did not understand all experimental details though; for example, on p. 38, isn’t a 1/3 probability to save “some” people trivially inferior to a certainty of saving “some” people? ratio bias: pp. 9-10 and 35 give references showing that people take 10:100 probability as higher than 1:10 probability, and that subjects reduce both probabilities and outcomes to categories. There is a nice comparison of the fuzzy-trace theory with the intuitionistic approach to mathematics of Brouwer. %} Reyna, Valerie F. & Charles J. Brainerd (1995) “Fuzzy-Trace Theory: An Interim Synthesis,” Learning and Individual Differences 7, 1–75. {% Measure risk attitudes by EU utility fitting (the Holt & Laury 2002 method), by an Eckel & Grossman method, and by psychometric questionnaire, among French farmers. The measures are correlated but not identical. Violations of EU can contribute to explaining the difference, as the authors note although still using EU à la Holt-Laury to fit data. The authors’ main conclusion is, then, that risk attitude is context dependent. A conclusion often favored by psychologists. %} Reynaud, Arnaud & Stéphane Couture (2012) “Stability of Risk Preference Measures: Results from a Field Experiment on French Farmers,” Theory and Decision 73, 203–221. {% Uses Tradeoff method to evaluate the assessment of mortality risks. %} Rheinberger, Christophe (2009) “Experimental Evidence against the Paradigm of Mortality Risk Aversion,” {% foundations of probability %} Rice, Adrian & Eugene Seneta (2005) “De Morgan in the Prehistory of Statistical Hypothesis Testing,” Journal of the Royal Statistical Society A 168, 615–627. {% Introduced idea of multiattribute risk aversion that plays role in Arne & I paper ACM model independently of his predecessor de Finetti (1932). %} Richard, Scott F. (1975) “Multivariate Risk Aversion, Utility Independence, and Separable Utility Functions,” Management Science 22, 12–21. {% Argues against the SG as gold standard for utility measurement because, first, EU is empirically violated (I agree) and, second, EU is neither appropriate normatively (I disagree) (SG doesn’t do well). He prefers the TTO. I agree with virtually all of pages 7-10, in particular that the author emphasizes that the SG cannot be a gold standard in view of violations of EU. I disagree more often with the texts following p. 10. The footnote on p. 11 cites in an affirmative manner the, I think incorrect, criticisms of Allais and Pope on the mathematics of Machina. P. 8, 2nd column, end of 1st para (referring to Gescheider 1988 for it): “As with other psychological concepts these attributes cannot be directly observed but only inferred. The concept itself is a construct and the functional relationship between the construct and external evidence must be embodied in psycho-physical theory.” P. 8 2nd column at about 2/3 of the page, on the ordinalist move in economics: “While removing the psychological connotations, this also reduced the value of the concept outside the framework of positive economics.” risky utility u = strength of preference v (or other riskless cardinal utility, often called value): p. 9, 2nd column: “It is likely that the great appeal of N-M utility in the context of CUA [Cost-Utility Analysis] is derived from such a conflation of concepts [representational utility versus strength of preference].” P. 10 discusses Utility of gambling (later the term utility of risk is also used). For the author, however, it seems to entail regret etc., any global aspect that cannot be modeled through the utility of single outcomes. P. 13 has a nice citation of Claude Bernard, taken from Allais. P. 18 discusses the HYE in a critical manner. %} Richardson, Jeff (1994) “Cost Utility Analysis: What Should Be Measured?,” Social Science and Medicine 39, 7–20. {% %} Richardson, Jeff, Jane Hall & Glenn Salkfeld (1996) “The Measurement of Utility in Multiphase Health States,” International Journal of Technology Assessment in Health Care 13, 35–48. {% questionnaire versus choice utility: measure choice utility through the HUI (which is based on EU for risk) and experienced utility through 5 introspective measures including EQ-5D, relate them, and find relations but not clear. Argue for nonlinear transformations to transform one into the other. %} Richardson, Jeff Richardson, Munir A. Khan, Angelo Iezzi, & Aimee Maxwell (2015) “Comparing and Explaining Differences in the Magnitude, Content, and Sensitivity of Utilities Predicted by the EQ-5D, SF-6D, HUI 3, 15D, QWB, and AQoL-8D Multiattribute Utility Instruments,” Medical Decision Making 35, 276–291. {% A German poet, often called (Jean) Paul wrote the following, a nice statement of loss aversion suggesting that it exceeds 2: Der Besitz macht uns nicht halb so glücklich, wie uns der Verlust unglücklich macht. (My translation: possession does not make us half as happy as loss makes us unhappy.) He lived from 1763 to 1825. %} Richter, Johann Paul Friedrich (17/18) {% revealed preference; This beautiful paper is the first to give completely necessary and sufficient conditions for revealed preference to be representable by a weak order, being an acyclicity condition, called congruency, in its Theorem 1. The term congruency, as the term rational, is not very informative. Many credit Varian (1982) for this result. The paper is a case of dillution: Theorem 1 is the most important result in all of revealed preference theory. All the rest in this paper is minor. %} Richter, Marcel K. (1966) “Revealed Preference Theory,” Econometrica 34, 635–645. {% revealed preference %} Richter, Marcel K. (1971) “Rational Choice.” In John S. Chipman, Leonid Hurwicz, Marcel K. Richter, & Hugo F. Sonnenschein (eds.) Preferences, Utility, and Demand, 29–58, Hartcourt, New York. {% %} Richter, Marcel K. (1975) “Rational Choice and Polynomial Measurement Theory,” Journal of Mathematical Psychology 12, 99–113. {% %} Richter, Marcel K. (1980) “Continuous and Semi-continuous Utility,” International Economic Review 21, 293–299. {% This paper is written in the spirit of Richter’s work, understanding very well how theoretical concepts should be related to observations and that deriving concepts from finitely many observed preferences is the thing to do. It shows how, under subjective expected utility with both utility and probability unknown, finitely many observations can reveal the info that subjective probabilities are in some interval [a,b] for any algebraic numbers a,b, and similar things. Algebraic means the solution to a polynomial equation with only natural numbers as weights involved. So we can find out that p1 is 2/3 or that it is squareroot of 2. We cannot find out that it is pi. At most we can find out that it is close to pi. Nice examples are given to illustrate this. Unfortunately, there are some advanced results on necessary and sufficient conditions for polynomial sets for which always utilities can be found and more like that which I did not find very interesting. %} Richter, Marcel K. & Leonard Shapiro (1978) “Revelations of a Gambler,” Journal of Mathematical Economics 5, 229–244. {% How to solve infinitely many linear inequalities. Probably related to Jaffray (1974). %} Richter, Marcel K. & Kam-Chau Wong (2004) “Infinite Inequality Systems and Cardinal Revelations,” Economic Theory 26, 947–971. {% The experiment uses hypothetical choice, because for environmental risks this is the only way, and then for best comparison also for financial. Extra pro is that financial choices then can use high significant amounts, where utility can be nonlinear for real reasons. Nicely, the author finds that Porsche club of America members do EU throughout, and elite rock climbers do so for financial risks. Measures probability weighting (as Tanaka et al. (2010 AER) for both financial and environmental risks. Confirms inverse-S (inverse-S). Probability overweighing of best outcomes is the same for financial and environmental, but for worst outcomes it is more pronounced for environmental. %} Riddel, Mary (2012) “Comparing Risk Preferences over Financial and Environmental Lotteries,” Journal of Risk and Uncertainty 45, 135–157. {% Use 2nd -order probability to model ambiguity, with normal distribution and variance reflecting ambiguity, and use it to quantitatively analyze an application of nuclear waste. %} Riddel, Mary & W. Douglass Shaw (2006) “A Theoretically-Consistent Empirical Model of Non-Expected Utility: An Application to Nuclear-Waste Transport,” Journal of Risk and Uncertainty 32, 131–150. {% Use two choicelists per person to derive two indifferences and then calculate two parameters, one the power of power-utility, the other one the inverse-S parameter of Prelec’s (1998) one-parameter family, which is taken to reflect the overweighting of small probabilities. Measure these for amateur car racers, technical rock climbers, SCUBA divers, and a student control group. Amateur auto racers are more rational in the sense of less probability weighting. Women, older subjects, and rock climbers transform probabilities more. As outcome the authors do not take money but life duration. They suggest that there have not been many measurements of utility of life duration, but there have been many in the health domain, including papers by my colleagues Attema and Bleichrodt. Unfortunately, the authors use the term risk aversion for concave utility, which is not correct under prospect theory, and the term multiple choice list, where multiple is redundant. In the choice situations, prospects are compared that have different outcomes but also different probabilities, which is not easy for subjects. %} Riddel, Mary & Sonja Kolstoe (2013) “Heterogeneity in Life-duration Preferences: Are Risky Recreationists Really More Risk Loving,” Journal of Risk and Uncertainty 46, 191–213. {% HYE Points out difference between continuous and discrete health flows in the debates; that CEs (certainty equivalents) are more naturally in terms of life years (for natural continuum) than in terms of health status and some other points. Some criticisms are not correct, e.g. in Footnote 50 on Johannesson, Pliskin & Weinstein 1993, because they refer, !in Ried’s terminology!, to HYE and not HYE-approach. %} Ried, Walter (1998) “QALYs versus HYEs—What’s Right and What’s Wrong. A Review of the Controversy,” Journal of Health Economics 17, 607–625. {% Does backward induction with multiple priors in maxmin EU sense. Then usually submartingales. Uses condition called rectangularity by Epstein & Schneider (2003, JET) that was also given by Sarin & Wakker (1998, JRU) and that is needed to have multiple priors as conjugate family. %} Riedel, Frank (2009) “Optimal Stopping with Multiple Priors,” Econometrica 77, 857–908. {% Show that for many prospects (lotteries) the measures of Aumann & Serrano (2008) and Foster & Hart (2009) are not defined because of divergence. Show that it is usually identical to or close to worst outcome. %} Riedel, Frank & Tobias Hellmann (2015) “The Foster-Hart Measure of Riskiness for General Gambles,” Theoretical Economics 10, 1–9. {% Games where players can choose to randomize using unknown probabilities (through Ellsberg urns provided to them), modeled using contraction EU of Gajdos et al. (2008). They use the term Ellsberg equilibria for the new equilibria. The data of Holt & Goereee (2001) can be accommodated by Ellsberg equilibria. %} Riedel, Frank & Linda Sass (2014) “Ellsberg Games,” Theory and Decision 76, 469–509. {% Show that probability estimates (judged probabilities, not decision-based, let be incentivized) of elements of a partition usually add to more than 1 also within-individually. More numerate subjects violated additivity less, especially if primed with numerical task first. (cognitive ability related to likelihood insensitivity) Direct matching, where subjects just directly choose probabilities, generates fewer additivity violations than when they choose from pre-chosen answer categories. %} Riege, Anine H. & Karl Halvor Teigen (2013) “Additivity Neglect in Probability Estimates: Effects of Numeracy and Response Format,” Organizational Behavior and Human Decision Processes 121, 41–52. {% %} Rieger, Marc Oliver (2017) “Comment on Cenci et al. (2015): “Half-Full or Half-Empty? A Model of Ddecision Making under Risk,” Journal of Mathematical Psychology 81, 110–113. {% A prospect over gains with finite expectation has finite expected utility if U is concave, but then need not have finite PT due to the overweighting of the high outcomes. Conditions about it are derived. Fig. 1 shows that w of T&K’92 need not be nondecreasing for = 0.2, and p. 668 gives formulas and details. P. 677 proposes w(p) = p + (3 – 3b)(p3 – (a+1)p2 + ap)/(a2–a+1) with 0 < a < 1 and 0 < b < 1 as new parametric family of weighting functions, with a the intersection with the diagonal (w(a) = a) and b a curvature parameter. They argue that this is the simplest polynomial with such a concave-convex switch. %} Rieger, Marc O. & Mei Wang (2005) “Cumulative Prospect Theory and the St. Petersburg Paradox,” Economic Theory 28, 665–679. {% Extend the separable Edwards version of prospect theory, with a normalization of weights, to continuous distributions. For each continuous distribution they choose one of several possible ways to approximate it discretely, and then define its value as the limit of the discrete approximations. In this way, the value of the continuous distribution depends only on probability weighting w through the derivative of w at 0. This convinces me that the model is not valuable for continuous distributions. It is a virtue of this paper to bring this point to the fore. %} Rieger, Marc Oliver & Mei Wang (2008) “Prospect Theory for Continuous Distributions,” Journal of Risk and Uncertainty 36, 83–102. {% %} Rieger, Marc Oliver & Mei Wang (2008) “What Is Behind the Priority Heuristic? A Mathematical Analysis and Comment on Brandstätter, Gigerenzer, and Hertwig (2006)” Psychological Review 115, 274–280. {% Used data from as in other studies by these authors, e.g. Rieger, Oliver, Wang, & Hens (2015 Management Science). Here students from many countries were asked a variation of Ellsberg's 3-color urn, where there are 30 red balls and 70 black or yellow balls. The most ambiguity averse country was Thailand (80% choose Red), and the last was the US (42% or so choose Red). They correlated these percentages with equity premiums in the countries, finding correlation 0.5 (p = 0.008). Macro-economic controls do not affect the result. Problem: their question did not control for suspicion (suspicion under ambiguity) and hence it may have been suspicion rather than ambiguity aversion that drove the correlation. They also correlated with Hofstede's (2001) uncertainty aversion index. It was positively correlated with ambiguity aversion, and explained the same variance in the equity premium puzzle as ambiguity aversion. %} Rieger, Marc Oliver & Mei Wang (2012) “Can Ambiguity Aversion Solve the Equity Premium Puzzle? Survey Evidence from International Data,” Finance Research Letters 9, 63–72. {% %} Riella, Gil (2015) “On the Representation of Incomplete Preferences under Uncertainty with Indecisiveness in Tastes and Beliefs,” Economic Theory 58, 571–600. {% The probabilistic dominance model works as follows. It is a regular Anscombe-Aumann framework. Let (A,f) be a set of acts containing f, where f has a special role: it is a status quo. The decision maker deemes as unacceptable all acts in A that have a probability of or more of yielding a utility loss relative to the status quo of or more. Here and are thresholds set by the decision maker. The unacceptable acts are removed from A. For the ones remaining, expected utility is maximized. A comparaitive condition of revealing more bias towards the status quo is defined (always having stronger preference for the status quo) which implies the same EU model but and being more extreme. %} Riella, Gil & Roee Teper (2014) “Probabilistic Dominance and Status Quo Bias,” Games and Economic Behavior 87, 288–304. {% Measure risk and ambiguity attitudes of 6912 subjects (students) in 53 countries, involving N=6912 students. Section 2 reviews other international studies, which never involved as many countries. Use WTP for gains but WTA for losses, doing hypothetical choice. Six gain lotteries and two loss lotteries, but no probability smaller than 0.1 or larger than 0.9, so cannot really observe inverse-S. Strictly speaking, the gain lotteries are not really gains because subjects pay their WTP, leading to net payment WTP (negative, so a loss) if the lottery gives outcome 0. Use as index of risk aversion the risk premium divided by the absolute value of EV. Because no mixed lotteries here and no EV = 0 this can be done, although, as is not well known, this normalization is too much and makes moderate payments too risk neutral. An analysis of these data determining PT parameters is in the authors’ 2017 paper in Theory and Decision. For ambiguity have 30 of 100 balls red, and the other 70 black or yellow in unknown proportion. 4.1% of the questions violate weak internality, and 15.1% strict. Risk averse for gains, risk seeking for losses: is found in all 53 countries. Positively related to Hofstede’s uncertainty avoidance index. gender differences in risk attitude: p. 642 §4.2.1: women are more risk averse for gains and more risk seeking for losses. Pp. 642-643: older people are less risk averse both for gains and for losses. P. 642: for gains, risk aversion is increasing in wealth between countries. Given that the index that the authors is more a relative risk aversion index than an absolute one, this is consistent with common findings at the individual level. For losses it is not significant (p. 643). reflection at individual level for risk: risk aversion for gains is negatively correlated with risk aversion for losses (p. 643). P. 645: using only students reduces heterogeneity within countries, making between-country comparisons more reliable. For 48 of 53 countries they have only one university. It is in itself good, if studying between-country variations, to have within-country homogeneity. Yet here typicalities of one particular university can much interfere with characteristics of the country. %} Rieger, Marc Oliver, Mei Wang, & Thorsten Hens (2015) “Risk Preferences around the World,” Management Science 61, 637–648. {% The authors published on this data set in Management Science in 2015, using a-theoretical indexes of risk attitudes such as normalized risk premium. This paper calculates five PT parameters, the same as T&K92, and then re-analyzes. The data of such a big study have to be noisy, and with eight questions per subject it is difficult to estimate five parameters of PT. Hence, they mostly take all answers per country assuming representative agent. One difficulty in this study is that for losses they only have prospects with one nonzero outcome, so that a common power of utility and probability weighting is unidentifiable. (Pointed out by the authors on p. 584.) Because the authors use a weighting function family, the one-parameter family of T&K92, their data fitting gives a unique fit, but this is due to assumed functions and not based on data. For gains they have only one of six prospects with more than one nonzero outcome, which should fully determine the power. gender differences in risk attitude: women do more probability weighting than men. concave utility for gains, convex utility for losses: is found (p. 582). Utility for losses is more linear than for gains, but not much. inverse-S: is found for both gains and losses. But they only fit the one-parameter family of TK92. Closer to linear for losses than for gains (p. 583). p. 583: utility parameters are related to portfolio decisions, but probability weighting parameters are not. This fits with my hypothesis that probability weighting is more noisy than utility. reflection at individual level for risk: p. 584 finds it, with a positive correlation between concavity of utility for gains and convexity for losses. P. 587: their nonparametric analysis of probability weighting depends much on utility assumed to be logpower, because only then the third displayed equation implies a constant ratio of CEs. P. 587: For losses, unlike for gains, the probability weighting parameter is not correlated with the nonparametric estimate, showing that the measurement for losses is more noisy than for gains. Of course, they have fewer observations for losses. P. 589: Of Hofstede’s indexes, individualism and uncertainty avoidance enhance more probability weighting. P. 593: Cites Hofstede (2001) on desirability, if studying between-country differences, to have within-country homogeneity of the sample. %} Rieger, Marc Oliver, Mei Wang & Thorsten Hens (2017) “Estimating Cumulative Prospect Theory Parameters from an International Survey,” Theory and Decision 82, 567–596. {% %} Riesbeck, Christopher K. & Roger C. Schank (1989) “Inside Case-Based Reasoning.” Lawrence Erlbaum, Hillsdale, NJ. {% P. 631 2nd column clearly specifies the topic of this paper: paternalism/Humean-view-of-preference: “Many have argued (e.g., Gerd Gigerenzer 1996a) that consistency principles are insufficient for defining rationality. If the achievement of an individual’s goal does not imply consistency, it is questionable whether functional behavior that violates consistency principles should be called “irrational.” ” Another cite is p. 632: “In contrast, we are interested in consistency principles that go beyond assumptions about the properties or attributes of the choice objects. For example, the transitivity axiom is applicable to a wide range of choice objects, …” %} Rieskamp, Jörg, Jerome R. Busemeyer, & Barbara A. Mellers (2006) “Extending the Bounds of Rationality: Evidence and Theories of Preferential Choice,” Journal of Economic Literature 44, 631–661. {% From abstract; Considers EU, PT, and decision field theory (DFT), in deterministic and probabilistic versions. The latter fit better than the former, and DFT does best. %} Rieskamp, Jörg (2008) “The Probabilistic Nature of Preferential Choice,” Journal of Experimental Psychology. Learning, Memory, and Cognition 34, 1446–1465. {% %} Riesz, Marcel (1927) “Sur les Maxima des Formes Bilinéaires et sur les Fonctionnelles Linéaires,” Acta Mathematica 49, 465–497. {% Use belief functions: and their updating is used to explain investment bubbles. The belief functions are not endogenous but exogenous, as in Jaffray’s works. They use Shafer’s 1976 updating. %} Rigotti, Luca, Matthew Ryan, & Rhema Vaithianathan (2016) “Throwing Good Money after Bad,” Decisions in Economics and Finance 39, 175–202. {% %} Rigotti, Luca & Chris Shannon (2005) “Uncertainty and Risk Aversion in Financial Markets,” Econometrica 73, 203–243. {% Take general convex preferences referring to Yaari (1969) for it and, as did the latter, take local marginal rates of substitution between states as kind of subjective probabilities or decision weights (can be interpreted as local beliefs). Show what this does in all kinds of models for ambiguity. Footnote 13 points out an inaccuracy in the proof of Billott, Chateauneuf, Gilboa, & Tallon (2000). Pp. 11790-1180 reminds me of a famous observation of Wald of the 1950s that a Pareto-optimal choice maximizes an expected value (through hyperplane supporting at optimum) which generates subjective probabilities. %} Rigotti, Luca, Chris Shannon, & Tomasz Strzalecki (2008) “Subjective Beliefs and ex Ante Trade,” Econometrica 76, 1167–1190. {% On “pariteitsschending,” meaning that left and right are not always symmetric in nature. %} Rikker, Geert & … (2000) {% Z&Z: shows that adverse selection can be detrimental for competitive markets. %} Riley, John G. (1979) “Informational Equilibria,” Econometrica 47, 331–359. {% Incompleteness in markets can be explained by ambiguity aversion. %} Rinaldi, Francesca (2009) “Endogenous Incompleteness of Financial Markets: The Role of Ambiguity and Ambiguity Aversion,” Journal of Mathematical Economics 45, 880–901. {% Generalize results on existence and continuity of solutions to Koopmans’ recursive equation. Consider consumption streams that have their growth rate unbounded above and below. %} Rincón-Zapatero, Juan Pablo & Carlos Rodríguez-Palmero (2007) “Recursive Utility with Unbounded Aggregators,” Economic Theory 33, 381–391.
Rippey, Robert M. & Anthony E. Voytovich (1983) “Linking Knowledge, Realism and Diagnostic Reasoning by Computer-Assisted Confidence Testing,” Journal of Computer-Based Instruction 9, 88–97. {% foundations of statistics: citing much on the debates. %} Risingener, D. Michael (2013) “Reservations about Likelihood Ratios and Some Other Aspects of Forensic ‘Bayesianism’,” Law, Probability and Risk 12, 63–73. {% conservation of influence %} Risjord, Mark (2005) “Reasons, Causes, and Action Explanation,” Philosophy of the Social Sciences 35, 294‑306. {% %} Riskey, Dwight R. & Michael H. Birnbaum (1974) “Compensatory Effects in Moral Judgments: Two Rights Don’t Make up for a Wrong,” Journal of Experimental Psychology 103, 171–173. {% People don’t want to vaccinate their child even if that decreases the total probability of death of the child, only so as to avoid perceived responsibility. %} Ritov, Ilana & Jonathan Baron (1990) “Reluctance to Vaccinate: Omission Bias and Ambiguity,” Journal of Behavioral Decision Making 3, 263–277. {% %} Ritov, Ilana & Jonathan Baron (1995) “Outcome Knowledge, Regret, and Omission Bias,” Organizational Behavior and Human Decision Processes 64, 119–127. {% %} Ritov, Ilana, & Daniel Kahneman (1997) “How People Value the Environment: Attitudes vs Economic Values.” In Max H. Bazerman, David Messick, Ann Tembrunzel, & Kimberly A. Wade-Benzoni (eds.) Psychological Approaches to Environmental and Ethical Issues in Management, New Lexington Press. {% game theory for nonexpected utility %} Ritzberger, Klaus (1996) “On Games under Expected Utility with Rank Dependent Probabilities,” Theory and Decision 40, 1–27. {% %} Ritzberger, Klaus (2008) “On Ranking of Journals in Economics and Related Fields,” Games and Economic Behavior 9, 402–430. {% Considers (and rejects) Fisher as inductive, says NP are deductive. Argues that these all are decision-theories. foundations of statistics %} Rivadulla, Andrés (1991) “Mathematical Statistics and Metastatistical Analysis,” Erkenntnis 34, 211–236. {% %} Rivero, J. Carlos, David R. Holtgrave, Robert N. Bontempo, & William P. Bottom (1989) “The St. Petersburg Paradox: Data, at Last,” Commentary 8, 46–51. Reprinted in Wing Hong Loke (ed.) Perspectives on Judgment and Decision Making, Lanham Press, Kent, England. {% Nice data illustrating loss aversion. For young male physicians between 1986 and 1990, the growth of income can best be explained through a model of reference points and loss aversion. %} Rizzo, John A. & Richard J. Zeckhauser (2004) “Reference Incomes, Loss Aversion, and Physician Behavior,” Review of Economics and Statistics 85, 909–922. {% %} Robert, Christian P. (1994) “The Bayesian Choice, A Decision-Theoretic Motivation; From Decision-Theoretic Foundations to Computational Implementation.” Springer, Berlin. (2nd edn. 2001.) {% Seems to have introduced the problem of the multi-armed bandit: a slot machine (one-armed bandit) may have more than one lever. When pulled, each lever provides a reward drawn from a distribution associated to that specific lever. The objective of the gambler is to maximize the collected reward sum through iterative pulls. It is classically assumed that the gambler has no initial knowledge about the levers, but through repeated trials, he can focus on the most rewarding levers. The exploration versus exploitation problem concerns to what extent one pulls the lever that performed best up to that time so as to maximize immediate reward, and to what extent one continues to pull levers inferior up to that point so as to continue collecting info about them. %} Robbins, Herbert E. (1952) “Some Aspects of the Sequential Design of Experiments,” Bulletin of the American Mathematical Society 55, 527–535. {% risky utility u = transform of strength of preference v, latter doesn’t exist: seems to have been very influential on the ordinal revolution. P. 16 of 1937 edn. seems to define economics: “Economics is the science which studies human behavior as a relationship between ends and scarce means which have alternative uses.” Often credited for being one of the main initiators of the ordinal revolution. P. 85 seems to write, about economics: “… is capable of being set out and defended in absolutely non-hedonistic term [and has no] essential connection with psychological hedonism, or for that matter with any other branch of Fach-Psychology.” %} Robbins, Lionel (1932) “An Essay on the Nature and Significance of Economic Science.” MacMillan, London. {% %} Robbins, Lionel (1938) “Interpersonal Comparisons of Utility: A Comment,” Economic Journal 48, 635–641. {% foundations of statistics: book review of Mayo & Spanos (2012) “Error and the Growth of Experimental Knowledge.” University of Chicago Press, Chicago. %} Robert, Christian (2013) “Error and Inference: An Outsider Stand on a Frequentist Philosophy,” Theory and Decision 74, 447–461. {% %} Roberts, Fred S. (1979) “Measurement Theory” (Encyclopedia of Mathematics and its Applications, Vol. 7). Addison-Wesley, London. {% Pp. 332-335 list emotional reasons other than aversion to unknown probabilities that can underlie the Ellsberg paradox. In his, long, reply, Ellsberg agrees with this view. %} Roberts, Harry V. (1963) “Risk, Ambiguity, and the Savage Axioms: Comment,” Quarterly Journal of Economics 77, 327–336. {% %} Roberts, John M., Jr. (1990) “Modeling Hierarchy: Transitivity and the Linear Ordering Problem” Journal of Mathematical Sociology 16, 77–87. {% foundations of probability; foundations of quantum mechanics; foundations of statistics: discusses how Bayesian view on subjective probability as degree of belief can go together with the view of quantum mechanics that nature is random. %} Roberts, John T. (2013) “Chance without Credence,” British Journal for the Philosophy of Science 64, 33–59. {% P. 135 proposes loss aversion, i.e., the utility kink at zero! Does assume concave utility throughout. Referred to in Robertson (1954, footnote 4). That footnote suggests that Chapman (1912) preceded him, but Chapman only has parts of increasing marginal utility and not loss aversion. %} Robertson, Dennis H. (1915) “A Study of Industrial Fluctuation; An Enquiry into the Character and Causes of the So-Called Cyclical Movement of Trade.” P.S. King & Son ltd., London. {% risky utility u = strength of preference v (or other riskless cardinal utility, often called value). Author writes informally, is probably text of spoken lecture. Presents himself as not formally trained. Says that he believes in cardinal utility and diminishing marginal utility on the basis of introspection. He is one of the few to think so in those days. Does not give formal arguments but suggests strong intuition. E.g. p. 667 l. 15-18. P. 673 footnote 4 describes loss aversion! A reaction is by Friedman (1955). %} Robertson, Dennis H. (1954) “Utility and All What?,” Economic Journal 64, 665–678. {% %} Robinson, Abraham (1974) “Non-Standard Analysis; revised edn.” Elsevier, New York. {% adaptive utility elicitation; find that VAS performs badly. %} Robinson, Angela, Paul Dolan, & Alan Williams (1997) “Valuing Health Status Using VAS and TTO: What Lies behind the Numbers,” Social Science and Medicine 45, 1289–1297. {% risky utility u = transform of strength of preference v. Authors use Schwartz’s (1998) proposal to correct VAS scores by means of Parducci’s R-F model, which describes range- and frequency biases. Seems to work OK for VAS. Unfortunately there is also a negative message, i.e., relating it to SG scores does not give good results. Did qualitative interviews of participants asking how they had reasoned. The interviews suggest that participants do take the sure outcome in the SG as a reference point, confirming the suggestion of Hersey & Schoemaker (1985). %} Robinson, Angela, Graham Loomes, & Michael Jones-Lee (2001) “Visual Analog Scales, Standard Gambles, and Relative Risk Aversion,” Medical Decision Making 21, 17–21. {% %} Robinson, Angela & Anne Spencer (2006) “Exploring Challenges to TTO Utilities: Valuing States Worse than Dead,” Health Economics 15, 393–402. {% Measure indifferences (p:H1, 1p: perfect health) ~(q:H2, 1q: perfect health), derived using matching, so two outcomes with one being perfect health. Under EU, if we scale U(perfect health) = 0, then this readily gives proportions of U and thus entire U for all health states worse than perfect health. Health states worse than dead need no special treatment here. This method has the (uninformative) name “modified standard gamble.” The authors cite preceding papers using it. They add an analysis base on RDU using power weighting function. Point out that T&K92 family did not work well, finding mostly pessimism (p. 346 penultimate para). They find pessimism and, hence, utility is less low (concave) than EU would have it. A problem with RDU is that power may not identifiable, most clearly seen if we scale U(perfect ealth) = 0. %} Robinson, Angela, Anne Spencer, & Peter Moffatt (2015) “A Framework for Estimating Health State Utility Values within a Discrete Choice Experiment Modeling Risky Choices,” Medical Decision Making 35, 276–291. {% A pet could be in one of two locations. Children did not know, but could put food in one or two locations. If the location was to be determined in the future, they would put food in both locations. If the location had been determined in the past (but unknown to them) they would put it in one of the two locations. Thus they treat uncertainty in the physical world (physical uncertainty) differently than when in their own perspective of ignorance (epistemic uncertainty). %} Robinson, Elizabeth J., Martin G. Rowley, Sarah R. Beck, Dan J. Carroll, & Ian A. Apperly (2006) “Children’s Sensitivity to Their Own Relative Ignorance: Handling of Possibilities under Epistemic and Physical Uncertainty,” Child Development 77, 1642–1655. {% Discusses behavioral economics, and the degree to which it enhances paternalism or better informing consumers. Download 7.23 Mb. Share with your friends:
|