Bibliography
preferring streams of increasing income
Download 7.23 Mb.
|
preferring streams of increasing income: stream of salary and percentile rankings in class: they prefer rising outcome to constant high outcome (with same final outcome), and they prefer constant low outcome to falling outcome (with same final outcome) %} Hsee, Christopher K. & Robert P. Abelson (1991) “Velocity Relation: Satisfaction as a Function of the First Derivative of Outcome over Time,” Journal of Personality and Social Psychology 60, 341–347. {% %} Hsee, Christopher K., Robert P. Abelson, & Peter Salovey (1991) “The Relative Weighting of Position and Velocity in Satisfaction,” Psychological Science 2, 263–266. {% Evaluability hypothesis: attributes receive more weight when evaluated jointly than when evaluated separately, because separately people see no way to evaluate whereas jointly they have something to compare. This can play a role in the inter- versus intra-personal tests of the Ellsberg paradox. Paper discusses a new preference reversal based not on difference in evaluation scale, but on difference in evaluation mode (joint versus separate evaluation), citing papers that did it before. %} Hsee, Christopher K., Sally Blount, George F. Loewenstein, & Max H. Bazerman (1999) “Preference Reversals between Joint and Separate Evaluations of Options: A Review and Theoretical Analysis,” Psychological Bulletin 125, 576–590. {% Short survey of many biases that make people not choose what is best: 1. Prediction biases: impact bias, projection bias, distinction bias, memory bias, belief bias. 2. Failures to follow predictions: impulsivity, rule-based decisiosns, lay rationalism, medium-maximization. They also discuss interactions. %} Hsee, Christopher K. & Reid Hastie (2006) “Decision and Experience: Why Don’t We Choose What Makes Us Happy?” TRENDS in Cognitive Sciences 10, 31–37. {% %} Hsee, Christopher K. & Howard C. Kunreuther (2000) “The Affection Effect in Insurance Decisions,” Journal of Risk and Uncertainty 20, 141–159. {% %} Hsee, Christopher K. & Yuval Rottenstreich (2004) “Music, Pandas, and Muggers: On the Affective Psychology of Value,” Journal of Experimental Psychological: General 133, 23–30. {% Choices are hypothetical. Let subjects (students) express own preference between a risky and a riskless prospect, let them guess what an anonymous other person would prefer, and let them guess what their neighbor (a concrete other) would prefer. Subjects predict that abstract others are more risk seeking (both for gains and for losses), but concrete others are the same. A risk-as-feeling hypothesis is put forward to explain. It is that subjects perceive of their deviation from risk neutrality as a nontypical emotional point, less applying to neutral others. This works for losses because for losses they find, opposite to prospect theory’s prediction (not pointed out by the authors; ), more risk aversion than risk seeking (see their Figure 1B). This complicates the finding. If, as usual, people are risk seeking for losses, then risk-as-feeling and others being more risk seeking become contradictory and it is not clear from this paper what to expect then. They also consider, but discard, other explanations such as a stereo-type explanation (others are Americans and their stereo-type is, as the authors claim, that Americans are venturous and risk-taking), where then it is apparently assumed that the other is defined as a member of a particular group, being American here. P. 45 2nd para claims that people consider risk seeking to be an admirable property. But I expect that most people find risk aversion to be more appropriate. P. 47 penultimate para of 1st column writes: “Consistent with prospect theory (Kahneman & Tversky, 1979), participants were more risk seeking in the loss condition than in the gain condition” but it does not mention that, contrary to prospect theory, they find risk aversion for losses rather than risk seeking. In study 3 they try to incentivize the prediction of the other choice: students were paired, seated next to each other, and received $50 if they predicted their neighbor’s choice correctly (p. 51 1st para). However, this is not a good incentive because it encourages everyone to choose, not what one likes, but what one expects one’s neighbor to predict. A practical problem is that it also encourages cribbing and communication. P. 51 end of 3rd para writes that not every right prediction gets rewarded, contrary to the 1st para, but only for 2 students out of 141 students. So the expected value of this is about 66 cents. Also, for the abstract other, a right prediction of majority-preference was rewarded. %} Hsee, Christopher K. & Elke U. Weber (1997) “A Fundamental Prediction Error: Self-Others Discrepancies in Risk Preference,” Journal of Experimental Psychology: General 126, 45–53. {% %} Hsee, Christopher K. & Elke U. Weber (1999) “Cross-National Differences in Risk Preference and Lay Predictions,” Journal of Behavioral Decision Making 12, 165–179. {% equity-versus-efficiency: %} Hsu, Ming, Cédric Anen, & Steven R. Quartz (2008) “The Right and the Good: Distributive Justice and Neural Encoding of Equity and Efficiency,” Science 320, 1092–1095. {% Degree of ambiguity in choices correlates positively with particular parts of the brains. Complete ambiguity is Ellsberg urn, other extreme is known urn. Then there are questions about temperatures in other cities, which are in between in ambiguity. There is also a guessing game against a better-informed opponent. In studies of ambiguity a difficulty is always how to control for belief. That is, people should avoid the unknown-probability event not because they consider it to be less likely as every Bayesian ambiguity neutral expected utility maximizer would then do the same waty, but they should do it for other reasons unlike Bayesians. Unfortunately, this study does not control for level of beliefs. Thus, in the knowledge questions subjects may prefer betting on high temperature in New York to betting on unknown city not because of ambiguity aversion, but simply because they consider it to be more likely in New York. (They can choose to bet on or against so will bet where the event more likely than its complement is most likely.) In the informed opponent game it is even worse, because every ambiguity neutral Bayesian person and every person I can think of should rather play the uninformed opponent, then the probabilities simply being better. Ambiguity arouses the same effects as the opponent-game. People with a particular brain damage are risk- and ambiguity neutral (although accepting a null with 6 subjects does not mean much), so, what many including me I consider rational. The data in the electronic web companion is strange. Table S6 reports the parameters of risk and ambiguity aversion. For the card-deck data there is a clear majority of ambiguity seeking! (ambiguity seeking) This deviates from common findings in the literature and from suggestions in the main text (p 1681 bottom of 1st column describes ambiguity aversion for the card-deck as the usual thing; p. 1682 bottom of 1st column has a null not rejected which, given 12 (or 16 as in table S6?) subjects, is problematic). For the knowledge question there is a clear majority risk seeking, which is also weird. When fitting the source function (where in many cases I do not know how they got the input p for the ambiguous events) they use the power family with the power as index of ambiguity aversion. %} Hsu, Ming, Meghana Bhatt, Ralph Adolphs, Daniel Tranel, & Colin F. Camerer (2005) “Neural Systems Responding to Degrees of Uncertainty in Human Decision Making,” Science 310, 9 Dec., 1680–1683. {% %} Hsu, Ming, Chen Zhao, & Colin F. Camerer (2006) “Nonlinear Probability Weighting in the Brain,” CalTech. {% A theoretical study of the effect of risk attitude on a two-stage English premium auction. %} Hu, Audrey, Theo Offerman, & Liang Zou (2011) “Premium Auctions and Risk Preferences,” Journal of Economic Theory 146, 2420–2439. {% PT, applications: considers transportation-waiting time as outcome, for risk decisions. Tests EU versus weighted utility, RDU, and PT. Only PT provides a slight improvement in fit. In EU, EV does as well as power or exponential utility, so they take EV (linear utility). For RDU considers Prelec 1 and 2 parameter families, T&K 92 family, and Goldstein & Einhorn (1987; they cite Gonzalez & Wu (1999) for it. For PT they do rank-dependent, with reference point endogenously estimated, with 8.8 minutes the resulting best reference point, and the only one that brings significant improvement. Seems that here they assume no parametric weighting functions but, with gains and losses weighted differently, can take the weight of each probability as a different parameter. %} Hu, Guotao, Aruna Sivakumar, & John W. Polak (2012) “Modelling Travellers Risky Choice in a Revealed Preference Context: A Comparison of EUT and Non-EUT Approaches,” Transportation 39, 825–841. {% Nice empirical study on reference-dependence in choices for food with reference levels within attributes. %} Hu, Wuyang, Wiktor L. Adamowicz, & Michelle M. Veeman (2006) “Labeling Context and Reference Point Effects in Models of Food Attribute Demand,” American Journal of Agricultural Economics 88, 1034–1049. {% Using tail comonotonicity in constructing multivariate distribuitions. %} Hua, Lei & Harry Joe (2012) “Tail Comonotonicity: Properties, Constructions, and Asymptotic Additivity of Risk Measures,” Insurance: Mathematics and Economics 51, 492–503. {% %} Hua, Wenxiu (1988) “The Properties of some Non-Additive Measures,” Fuzzy Sets and Systems 27, 373–377. {% P. 16: robustness of EU results, demonstrated by Machina, is argument in favor of EU!! %} Huang, Chi-Fu & Robert H. Litzenberger (1988) “Foundations for Financial Economics.” North-Holland, Amsterdam. {% Preceding papers considered Choquet integrals of set-valued functions that were set-valued. This paper proposes one that is real-valued. %} Huang, Yan & Congxin Wu (2014) “Real-Valued Choquet Integrals for Set-Valued Mappings,” International Journal of Approximate Reasoning 55, 683–688. {% Christiane, Veronika & I: participants (“agents”) should maximize the utility for someone else (“principal”), which consists of aggregating three components (ski vacation with price, probability of snow, and quality of slope). They are told exactly how to aggregate the values of the separate components, through a weighted sum with both attribute weights and attribute values specified. Only, the values are not given numerically, but are indicated through points on a line (i.e., a kind of VAS score) without any ruler provided. So, the whole value system has been specified and only the numerical processing matters. The participants were first trained through 7 choice and 9 matching questions in the first experiment, and a few more in a second experiment, where they received rewards as they were closer to the true values. At the end, p. 88, the authors distinguish two steps in preference valuations: (1) Creating an internal representation of the information [values] and (2) expressing these representations through a specific task. I guess that, in our terminology, (1) refers to intrinsic value, (2) to, a.o., numerical sensitivity. There are three modes of response, matching, choice, and rating. The authors write in the “paternalistic” way that I like, where biases are things to be corrected for (paternalism/Humean-view-of-preference). They consider “negativity bias” (also called level focusing) which may be more general than loss aversion but with these data (three levels per attribute) is the same. Findings (pp. 86-87): Strong scale compatibility. No prominence effect, if anything, the opposite. - Choice: authors are happily surprised that the participants make compensatory tradeoffs among attributes, rather than resort to noncompensatory heuristics. There still is considerable loss aversion. - Ratings: take less than half of time of other modes of response, have about half the loss aversion of choice, noisier. - Matching: most difficult. Curvature of scale is best captured, no loss aversion (rationale on p. 70: matching pairs provides its own reference points), only problem is much scale compatibility. So, it’s good for relative comparisons of the nonmatching dimensions. P. 73, however, suggests that matching enhances looking only at differences of attribute, thus to “overlinearization” (may contribute to: CE bias towards EV). Note, however, that linear processing of attributes seems to be rational in this empirical study, given that these are already evaluations of attributes. Subjects judge that choice (not binary but always from triples) is best, then matching, last rating. IMPORTANT: as the authors remark on p. 88, 3rd para, their finding is important because it shows that loss aversion occurs not (merely) at the level of intrinsic values, but also is a bias in the process of expressing intrinsic values: “In many settings, one cannot tell whether loss aversion is a bias or merely a reflection of the fact that losses have more emotional impact than gains of equal magnitude. In our choice and rating tasks, however, we found clear evidence that agents motivated to accurately represent the preferences of others gave more weight to negative outcomes than is appropriate.” %} Huber, Joel, Dan Ariely, & Gregory Fischer (2001) “Expressing Preference in a Principal-Agent Task: A Comparison of Choice, Rating, and Matching,” Organizational Behavior and Human Decision Processes 87, 66–90. {% measure of similarity; context-dependence, violation of IIA, called “attraction effect” (or “asymmetric dominance”) where adding a dominated alternative increases choice percentage of chosen alternative à la Tversky & Simonson. Seems that this 1982 paper was the first. %} Huber, Joel, John W. Payne, & Christopher Puto (1982) “Adding Asymmetrically Dominated Alternatives: Violations of Regularity and the Similarity Hypothesis,” Journal of Consumer Research 9, 90–98. {% Subjects hypothetically judge quality of areas based on cost of living and quality of water in lakes and rivers. Reference dependence and loss aversion can clearly be generated by proper framing. In iterated choice, the first option offered and the last one before choosing now have much effect. %} Huber, Joel, W. Kip Viscusi, & Jason Bell (2008) “Reference Dependence in Iterative Choices,” Organizational Behavior and Human Decision Processes 106, 143–152. {% %} Huber, Joel, Dick R. Wittink, John A. Fiedler, & Richard Miller (1993) “The Effectiveness of Alternative Preference Elicitation Procedures in Predicting Choices,” Journal of Marketing Research 30, 105–114. {% %} Huber, Peter J. (1973) “The Use of Choquet Capacities in Statistics,” Bulletin de l’Institut International de Statistique 45, 181–191. {% %} Huber, Peter J. (1981) “Robust Statistics.” Wiley, New York. {% %} Huber, Peter J. & Volker Strassen (1973) “Minimax Tests and the Neyman-Pearson Lemma for Capacities,” Annals of Statistics 1, 251–263. {% homebias: seems to show that within same country there is a kind of homebias for own region. %} Huberman, Gur (2000) “Familiarity Breeds Investment,” Review of Financial Studies 46, 3–28. {% %} Hubert, Lawrence (1974) “Some Applications of Graph Theory and Related Non-Metric Techniques to Problems of Approximate Seriation: The Case of Symmetric Proximity Measures,” British Journal of Mathematical and Statistical Psychology 27, 133–153. {% %} Hubert, Lawrence (1974) “Problems of Seriation Using a Subject by Item Response Matrix,” Psychological Bulletin 81, 976–983. {% %} Hubert, Lawrence (1974) “Some Applications of Graph Theory to Clustering,” Psychometrika 39, 283–309. {% %} Hubert, Lawrence (1976) “Seriation Using Asymmetric Proximity Measures,” British Journal of Mathematical and Statistical Psychology 29, 32–52. {% %} Hubert, Lawrence & James Schultz (1976) “Quadratic Assignment as a General Data Analysis Strategy,” British Journal of Mathematical and Statistical Psychology 29, 190–241. {% Extend Koopmans to algebraic structure, first for bounded structures. Also present a result for unbounded structures but, correctly, point out in the Concluding Remarks that these conditions are not directly testable. Wakker’s (1993, MOR) truncation continuity would provide an alternative way to go here. %} Hübner, Ronald & Reinhard Suck (1993) “Algebraic Representation of Additive Structure with an Infinite Number of Components,” Journal of Mathematical Psychology 37, 629–639. {% In bargaining situations it can be advantageous to commit to the endowment effect. The authors derive evolutionary arguments for the endowment effect from this observation. %} Huck, Steffen, Georg Kirchsteiger, & Jörg Oechssler (2005) “Learning to Like What You Have—Explaining the Endowment Effect,” Economic Journal 115, 689–702. {% Do Allais paradox, with high hypothetical payoffs, with low hypothetical, and with low real. For representative CentER panel, and for student group. Find rather low violation rates for low payments. More violations in population than in student group. More violations for low-educated. %} Huck, Steffen & Wieland Müller (2012) “Allais for All: Revisiting the Paradox in a Large Representative Sample,” Journal of Risk and Uncertainty 44, 261–293. {% probability elicitation: applied to experimental economics; Use proper scoring rules (the quadratic rule) and the measurement of matching probabilities, derived from certainty equivalents using linear utility (eliciting certainty equivalents through BDM (Becker-DeGroot-Marschak)) to measure beliefs about percentages of strategy choices of other players in other games. The quadratic scoring rules are more accurate. Beliefs are conservative; i.e., biased towards 0.5 (e.g. p. 72 penultimate para). They did not give explanation about properness of the quadratic scoring rule, and just did it (p. 75 footnote 9), but just asked for probability judgment and applied the scoring rule. %} Huck, Steffen & Georg Weizsäcker (2002) “Do Players Correctly Estimate What Others Do? Evidence of Conservatism in Beliefs,” Journal of Economic Behavior and Organization 47, 71–85. {% utility = representational?: the whole issue no. 3 of Synthese is on coherentism. %} Huemer, Michael (2007) “Weak Bayesian Coherentism,” Synthese 157, 337–346. {% Consider two-outcome prospects. Ambiguity was generated as second-order probability, with reduced probabilities 0.25, 0.50, and 0.75, and in ambiguity always one outcome was 0. It was not explained to the subjects what the reduced probabilities under ambiguity would be but subjects saw several drawings so that after a while they could figure out a bit about different levels of likelihood when deciding under ambiguity. For risky choice they assumed EU with relative risk aversion indexing risk attitude. For ambiguity they took the utility function inferred from risky choice, and then did -maxmin for from [0,1]; i.e., the model (1)u(x1) + u(x2) with x1 > x2. It is the usual two-outcome RDU or prospect theory or biseparable model, or Arrow-Hurwicz model (extended to multiple priors by Ghirardato & Marinacci 2004) given that subjects cannot know, apparently, what the level of likelihood is under ambiguity. They find that different parts of the brain get activated under ambiguity than under risk (e.g. p. 772: [risk and ambiguity] “represent two types of decision making that are supported by distinct [brain] mechanisms.”. This is indirect evidence that risk and ambiguity are not related (correlation risk & ambiguity attitude). Although they have the data, they do not report relations between risk and ambiguity attitudes. %} Huettel, Scott A., C. Jill Stowe, Evan M. Gordon, Brent T. Warner, & Michael L. Platt (2006) “Neural Signatures of Economic Preferences for Risk and Ambiguity,” Neuron 49, 765–775. {% information aversion!!! If person is tested on HD (Huntington’s Disease), it is found out if person is risky (97% chance of getting HD), or not risky (3% chance of getting HD). But this is then also done for the members of the family. Then these members can, without any more trouble, get to know if they are risky or not. Similar, if child in mother is tested (to be aborted if risky) then mother is also tested and be informed. Often mothers prefer not to know about themselves. %} Huggins, Marlene, Maurice Bloch, Shelin Kanani, Oliver W.J. Quarrell, Jane Theilman, Amy Hedrick, Brnard Dickens, Abbyann Lynch, & Michael Hayden (1990) “Ethical and Legal Dilemmas Arising during Predictive Testing for Adult-Onset Disease: The Experience of Huntington Disease,” American Journal of Human Genetics 47, 4–12. {% Gotten from Palli Sipos; foundations of quantum mechanics %} Hughes, R.I.G. (1981) “Quantum Logic,” Scientific American 245 (No. 4), 146–157. {% %} Hughly, Philip & Charles Sayward (1990) “Can there Be a Proof That Some Unprovable Arithmetic Sentence is True?,” Dialectica 43, 289–292. {% finite additivity: %} Huisman, Leendert (2015) “Reflecting on Finite Additivity,” Synthese 192, 1785–1797. {% (ISBN: 0-13-149908-4) %} Hull, John C. (2005) “Options, Futures, and Other Derivatives.” Englewood Cliffs, Prentice-Hall, NJ (6th edn.). {% %} Hull, John C. (2006) “Options, Futures, and Other Derivatives: Solutions Manual.” (ISBN: 0-13-149906-8) Englewood Cliffs, Prentice-Hall, NJ. {% %} Hull, John C. (2013) “Options, Futures, and Other Derivatives.” Englewood Cliffs, Prentice-Hall, NJ (9th edn.). {% %} Hüllermeier, Eyke (2007) “Case-based Approximate Reasoning.” Springer, Berlin. {% conservation of influence: Part 1, 118 seems to write: “[t]here is implanted in the human mind a perception of pain and pleasure as the chief spring and moving principle of all its actions” paternalism/Humean-view-of-preference: Part 3 Of the will and direct passions, Sect. 3 Of the influencing motives of the will writes: “What may at first occur on this head, is, that as nothing can be contrary to truth or reason, except what has a reference to it, and as the judgments of our understanding only have this reference, it must follow, that passions can be contrary to reason only so far as they are accompany'd with some judgment or opinion. According to this principle, which is so obvious and natural, `tis only in two senses, that any affection can be call'd unreasonable. First, When a passion, such as hope or fear, grief or joy, despair or security, is founded on the supposition or the existence of objects, which really do not exist. Secondly, When in exerting any passion in action, we chuse means insufficient for the design'd end, and deceive ourselves in our judgment of causes and effects. Where a passion is neither founded on false suppositions, nor chuses means insufficient for the end, the understanding can neither justify nor condemn it. `Tis not contrary to reason to prefer the destruction of the whole world to the scratching of my finger. `Tis not contrary to reason for me to chuse my total ruin, to prevent the least uneasiness of an Indian or person wholly unknown to me. `Tis as little contrary to reason to prefer even my own acknowledge'd lesser good to my greater, and have a more ardent affection for the former than the latter. A trivial good may, from certain circumstances, produce a desire superior to what arises from the greatest and most valuable enjoyment; nor is there any thing more extraordinary in this, than in mechanics to see one pound weight raise up a hundred by the advantage of its situation. In short, a passion must be accompany'd with some false judgment. in order to its being unreasonable; and even then `tis not the passion, properly speaking, which is unreasonable, but the judgment.” Seems to have said (p. 413): “Reason alone can never be a motive to any action of the will;” (p. 415): “reason is, and ought only to be the slave of the passions.” “a passion must be accompay’d with some false judgment, in order to its being unreasonable.” %} Hume, David (1740) “A Treatise of Human Nature.” (1978, Clarendon Press, Oxford.) {% conservation of influence: at the end of the section that deals with causation, Hume states: “we may define cause to be an object followed by another, and where all the objects, similar to the first, are followed by objects similar to the second. Or, in other words, where, if the first object had not been, the second never had existed.” Second formulation is a difference-making idea. A cause makes a difference to whether its effect obtains: without it, the effect would not have obtained. %} Hume, David (1995) “An Inquiry Concerning Human Understanding.” Upper Saddle River: Prentice Hall. {% coalescing %} Humphrey, Steven J. (1995) “Regret Aversion or Event-Splitting Effects? More Evidence under Risk and Uncertainty,” Journal of Risk and Uncertainty 11, 263–274. {% coalescing %} Humphrey, Steven J. (1996) “Do Anchoring Effects Underlie Event-splitting Effects? An Experimental Test,” Economic Letters 51, 303–308. {% coalescing %} Humphrey, Steven J. (1998) “More Mixed Results on Boundary Effects,” Economics Letters 61, 79–84. {% coalescing %} Humphrey, Steven J. (1999) “Probability Learning, Event-splitting Effects and the Economic Theory of Choice,” Theory and Decision 46, 51–78. {% %} Humphrey, Steven J. (2000) “The Common Consequence Effect: Testing a Unified Explanation of Recent Mixed Evidence,” Journal of Economic Behavior and Organization 41, 239–263. {% coalescing %} Humphrey, Steven J. (2001) “Non-transitive Choice: Event-Splitting Effects or Framing Effects?,” Economica 68, 77–96. {% coalescing %} Humphrey, Steven J. (2001) “Are Event-Splitting Effects Actually Boundary Effects?,” Journal of Risk and Uncertainty 22, 79–93. {% %} Humphrey, Steven J. (2004) “Feedback-Conditional Regret Theory and Testing Regret Aversion in Risky Choice,” Journal of Economic Psychology 25, 839–857. {% coalescing; Investigates effects of learning on violations of monotonicity, coalescing, and the common consequence effects (more complex than Allais, with no sure option available). Certainty equivalents of prospects are elicited through matching. From that, choices are derived indirectly. In the learning treatment, subjects are shown 10 drawings of each prospect before deciding. These drawings were manipulated so as to be representative (deception when implementing real incentives). Some deviations from expected utility were reduced but others were enhanced. The author is, understandably, more enthusiastic about his own research speciality, event splitting, than about other topics when he writes (p. 97): “Event-splitting effects are unlike many choice anomalies because there exists a range of real world decision-making contexts where one observed analogous behaviour.” %} Humphrey, Steven J. (2006) “Does Learning Diminish Violations of Independence, Coalescing and Monotonicity?,” Theory and Decision 61, 93–128. {% PT falsified & inverse-S: they test the common consequence effect and find risk aversion increasing and not decreasing, which is the exact opposite of inverse-S. This independently replicates the same finding as by Birnbaum, for instance in Birnbaum & Chavez (1997). Use random incentive system. Did it with poor farmers from the countries mentioned in the title. More elaborate results, with error theories added, are in Humphrey & Verschoor (2004, Journal of African Economies). Unfortunately, the papers have no cross references to explain their overlap and priority. %} Humphrey, Steven J. & Arjan Verschoor (2004) “The Probability Weighting Function: Experimental Evidence from Uganda, India and Ethiopia,” Economics Letters 84, 419–425. {% PT falsified & inverse-S: do same as their 2004 Economics Letters paper, but more elaborate, with error theory added. Then still they prefer RDU with error better than EU with error. (e.g. p. 82 & 84) %} Humphrey, Steven J. & Arjan Verschoor (2004) “Decision-Making under Risk among Small Farmers in East Uganda,” Journal of African Economies 13, 44–101. {% conditional probability %} Humphreys, Paul (2004) “Some Considerations on Conditional Chances,” British Journal for the Philosophy of Science 55, 667–680. {% It is well known that nudging people into reducing energy use works well if social comparisons are brought in. This paper examines long-term effects. People slowly react to the nudge, only slowly reducing energy use, but after a prolonged exposure the effect remains long after. %} Hunt Allcott & Todd Rogers (2014) “The Short-Run and Long-Run Effects of Behavioral Interventions: Experimental Evidence from Energy Conservation,” American Economic Review 104, 3003–3037. {% dynamic consistency %} Huntley, Nathan & Matthias C.M. Troffaes (2012) “Normal Form Backward Induction for Decision Trees with Coherent Lower Previsions,” Annals of Operations Research 195, 111–134. {% Schijnt al IIA-versie gehad te hebben. %} Huntington, E.V. (1938) “A Paradox in the Scoring of Competing Teams,” Science 88, 287‑288. {% %} Hurd, Michael D. & Kathleen McGarry (2002) “The Predictive Validity of Subjective Probabilities of Survival,” Economic Journal 112, 966–985. {% %} Hurkens, Sjaak (1996) “Multi-Sided Pre-play Communication by Burning Money,” Journal of Economic Theory 69, 186–197. {% Seems to show that subjects like to answer truthfully, and not lie, also if no incentive. %} Hurkens, Sjaak & Navin Kartik (2009) “Would I Lie to You? On Social Preferences and Lying Aversion,” Experimental Economics 12, 180–192. {% Suggest that overbetting on outsiders and underbetting on favorites may be due to cost of information, and other things. So, variation on information-sensitivity. Their data do not find much, H0. %} Hurley, William & Lawrence McDonough (1995) “A Note on the Hayek Hypothesis and the Favorite-Longshot Bias in Parimutual Betting,” American Economic Review 85, 949–955. {% Put red and white poker chips in bag (actually, coffee can), say 5 red and 3 white, 8 in total. Then asked subjects to predict how many reds there would be in, say, 5 drawings, always with replacement. Subjects received a prize if they guessed right. Obviously, they should gamble on the most likely result of the five drawings. Seems that they did not do this very well, but for small real probability of red acted as if this probability was higher, and for large real probability as if it was smaller (inverse-S). I did not understand on p. 176 the discussions of work of Karni, first because for given probabilities state-dependence does not seem to be plausible, second, how they could escape from it if it would nevertheless arise. Conclude that previous conclusions in the literature about divergence of subjective and objective probabilities may be based on faulty assumptions, such as strict rationality. %} Hurley, Terrence M. & Jason F. Shogren (2005) “An Experimental Comparison of Induced and Elicited Beliefs,” Journal of Risk and Uncertainty 30, 169–188. {% %} Hurvich, Leo M. & Dorothea Jameson (1951) “Psychophysical Study of White. I. Neutral Adaptation,” Journal of the Optical Society of America 41, 521–527. {% event/utility driven ambiguity model: event-driven Introduced the -maxmin model in its Remark 4, a little below the displayed equation. %} Hurwicz, Leonid (1951) “Some Specification Problems and Applications to Econometric Models” (Abstract), Econometrica 19, 343–344. {% %} Hurwicz, Leonid (1951) “Optimality Criteria for Decision Making under Ignorance,” Cowles Commission Discussion Paper, Statistics, No. 370, mimeographed. {% %} Hurwicz, Leonid (1960) “Optimality and Informational Efficiency in Resource Allocation.” In Kenneth J. Arrow, Samuel Karlin, & Patrick Suppes (1960, eds.) Mathematical Methods in the Social Sciences, 17–46, Stanford University Press, Stanford, CA. {% Is credited by Nobel-2007 committee for having introduced incentive compatibility. Incentive compatibility occurred before in proper scoring rules by Brier (1950) and de Finetti (1962). %} Hurwicz, Leonid (1972) “On Informationally Decentralized Systems.” In Charles Bartlett McGuire & Roy Radner (eds.) Decision and Organization, 297–336, North-Holland, Amsterdam. {% revealed preference %} Hurwicz, Leonid & Marcel K. Richter (1971) “Revealed Preference without Demand continuity Assumptions.” In John S. Chipman, Leonid Hurwicz, Marcel K. Richter, & Hugo F. Sonnenschein (eds.) Ch. 3, “Preferences, Utility, and Demand,” Hartcourt, New York. {% P. 34 seems to define good through pleasure, an object is good if it creates pleasure. Looks already quite like utility. %} Hutcheson, Francis (1728) “An Essay on the Nature and Conduct of the Passions and Affections.” J. Osborne and T. Longman, London. {% %} Hutton Barron, Francis & Bruce E. Barrett (1996) “Decision Quality Using Ranked Attribute Weights,” Management Science 42, 1515–1523. {% Found high convergence between risky and riskless utility. Derive theoretical relations, if one is additive, the other is multiplicative, then, by Cauchy’s equation ... etc. Find that linear relation gives good fit. Exponential transform provides little gain. utility elicitation; risky utility u = transform of strength of preference v %} Hutton Barron, Francis, Detlof von Winterfeldt, & Gregory W. Fischer (1984) “Empirical and Theoretical Relationships between Value and Utility Functions,” Acta Psychologica 56, 233–244. {% Seems to formulate principle of expected value. Blaise Pascal seems to have encouraged him to write this book. %} Huygens, Christiaan (1657) “Tractatus de Ratiociniis in Ludo Aleane.” Amsterdam. Translated into Dutch by Frans van Schooten: Van Reeckening in Spelen van Geluck. {% %} Hwang, Ching Lai & Kwangsun Yoon (1981) “Multiple Attribute Decision Making.” Springer, Berlin. {% %} Hwang, Ching Lai, Abu S.M. Masud (1979) (in collaboration with Sudhar R. Paidy & Kwangsun Yoon) “Multiple Objective Decision Making: Methods and Applications: A State-of-the-Art Survey.” Springer, Berlin. {% %} Ibanez Marcela, Simon Czermak, & Matthias Sutter (2009) “Searching for a Better Deal. On the Influence of Group Decision Making, Time Pressure and Gender in a Search Experiment,” Journal of Economic Psychology 30, 1–10. {% time preference: comparing risky and intertemporal utility: let people do hypothetical choices between payments (one nonzero) with both risk and delay, assume constant discounting and EU with CRRA, and fit the parameters simultaneously. They call this new but Andersen et al. (2008 Econometrica) and Chapman (1996) and others preceded them. Correlate their findings with smoking behavior. %} Ida, Takanori & Rei Goto (2009) “Simultaneous Measurement of Time and Risk Preferences: Stated Preference Discrete Choice Modeling Analysis Depending On Smoking Behavior,” International Economic Review 50, 1169–1182. {% P. 244: “It seems wiser to treat numerical estimates of chance as behavioral indicators of underlying evidence.” [italics from original] Give arguments favoring qualitative rather than quantitative expressions of uncertainty. %} Idson, Lorraine Chen, David H. Krantz, Daniel Osherson, & Nicolao Bonini (2001) “The Relation between Probability and Evidence Judgment: An Extension of Support Theory,” Journal of Risk and Uncertainty 22, 227–249. {% Big study in Japan finds that discounting, also hyperbolic, is related to body weight. Natural that obesity and the like will be related to this. Sign dependence is also related to it. %} Ikeda, Shinsuke, Myong-Il Kang, & Fumio Ohtake (2010) “Hyperbolic Discounting, the Sign Effect, and the Body Mass Index,” Journal of Health Economics 29, 268–284. {% one-dimensional utility: Pareto utility is power utility with initial wealth incorporated. The author discuss advantages of this family. %} Ikefuji, Masako, Roger J. A. Laeven, Jan R. Magnus, & Chris Muris (2013) “Pareto Utility,” Theory and Decision 75, 43–57. {% Application of ambiguity theory; Studies financial markets, with optimal portfolios and equilibrium asset prices, and the effects of ambiguity aversion as in maxmin EU of Gilboa & Schmeidler (1989). The implied desire to hedge leads to portfolio inertia (also if free market, and also for investors who do participate in the market). Small pieces of news can lead to drastic changes and excess volatility. Interaction between risk and ambiguity may explain spikes in stock price volatility. %} Illeditsch, Philipp Karl (2011) “Ambiguous Information, Portfolio Inertia, and Excess Volatility,” The Journal of Finance 66, 2213–2247. {% %} Incekara, Elif & Jack Stecher (2015) “An Experimental Test of Theories of Behavior in Allais-Type Tasks,” working paper. {% This paper criticizes nudging techniques because advocates (including me) assume the existence of true correct best values. They assume that there is “something down there” (my words). The authors argue that this assumption is unfounded. E.g. p. 13: “Thus, Hausman’s analysis does not resolve the problem we identified in the literature of behavioural welfare economics. That problem was to justify the implicit assumption that, for any given individual, there exists some mode of latent reasoning that generates complete and context-independent subjective preferences.” P. 22 (conclusion): “We need a normative economics that does not presuppose a kind of rational human agency for which there is no known psychological foundation.” The paper often cites Kahneman as an authority. It takes space to put every possible detail right. P. 1 1st para describes what I call the Bayesian twin, although here it is broader: “The task for welfare economics is then to reconstruct the preferences that the individual would have acted on, had her reasoning not been distorted by whatever psychological mechanisms were responsible for the mistakes, and to use the satisfaction of these reconstructed preferences as a normative criterion.” The paper sometimes calls that “preference purification” (title of §2 and elsewhere). P. 2 3rd para: “Although there is a clear sense in which the choices made (or preferences revealed or judgements expressed) by the person in different contexts are inconsistent with one another, it is not at all obvious which (if any) of these choices is correct – or even how ‘correctness’ should be defined.” P. 7, on the often cited Bernheim & Rangel (2007, 2009): “Bernheim and Rangel’s first line of approach is to propose a criterion that respects the individual’s revealed preferences over pairs of objects if those preferences are not affected by changes in ancillary conditions, and instructs the planner ‘to live with whatever ambiguity remains’ (2009, p. 53). They then suggest that this rather unhelpful criterion might be given more bite by the deletion of ‘suspect’ GCSs. A GCS is deemed to be suspect if its ancillary conditions induce impairments in the individual’s ability to attend to or process information or to implement desired courses of action.” [Italics added] P. 13 cites Hausman & Welch (2010 p. 128) pointing out that nudge does not fully 100% respect free will: “something paternalistic, not merely beneficent …in addition to or apart from rational persuasion, they may ‘push’ individuals to make one choice rather than another … their autonomy – the extent to which they have control over their own evaluations and deliberations – is diminished. Their actions reflect the tactics of the choice architect rather than exclusively their own evaluation of alternatives. … limiting what choices are available or shaping choices risks circumventing the individual’s will. (p. 130)” Infante et al. call the Bayesian twin the “inner rational agent.” P.. 14: “We will call this disembodied entity the inner rational agent. … Preference purification can be thought of as an attempt to reconstruct the preferences of the inner rational agent by abstracting from the distorting effects of – by ‘seeing through’ – the psychological shell. … if the faults in the psychological shell were corrected. Several parts discuss Bleichrodt, Pinto, & Wakker (2001), abbreviated BPW, in interesting manners. Little surprise that I disagree sometimes, in two places. The first is p. 20: “BPW’s purification methodology treats the non-linearity of the probability weighting function as a reasoning error … But where is the error? … had used decision weights in the mistaken belief that they were objective probabilities. But that is not a remotely plausible account … remember that when people respond to Allais’ problems, they are told all the relevant objective probabilities.” This discussion interprets probability weighting too narrowly. It need not just be wrong cognitive belief about probability. It can also be wrong FEELING while right knowing (imperfect numerical sensitivity), or pessimistic overattention to worst outcomes, or deliberate nonlinear decision weighting, e.g. by researchers who think that nonEU for risk is rational (which I Bayesian then still consider to be a mistake to be corrected for). The overly narrow interpretation of probability weighting here is called the second misunderstanding in Fox, Erner, & Walters (2015 p. 55). P. 21 3rd para, middle of page, writes that BPW would not go by the preferences of the subject but by those of the professional: “Viewed in this way, what seems to be required is not an inference about the hypothetical choices of the client’s inner rational agent, but rather a way of regularising the available data about the client’s preferences so that it is compatible with the particular model of decisionmaking that the professional wants to use.” However, BPW assume as default that the only thing the professional wants to do is maximize the subject’s preferences. The professional does not have an own agenda. P. 21 penultimate para goes a long way agreeing with BPW, despite the (“religious”) difference in view on the existence of true values: “In the same way, a medical decision-maker might reasonably use BPW’s methodology to construct a tractable model of the client’s preferences, regularised so as to be consistent with expected utility theory, without claiming that the preferences in the model were latent in the client. The arguments we have developed in this paper would not be objections to a version of behavioural welfare economics that claimed only to regularise revealed preferences that were inconsistent with conventional theory, without interpreting this process as the identification and correction of errors, or as a way of helping individuals to make better choices. But that is not the version of behavioural welfare economics that is to be found in the literature.” %} Infante, Gerardo, Guilhem Lecouteux, & Robert Sugden (2016) “Preference Purification and the Inner Rational Agent: A Critique of the Conventional Wisdom of Behavioural Welfare Economics,” Journal of Economic Methodology 23, 1–25. {% anonymity protection %} International Journal of Uncertainty, Fuzziness & Knowledge-Based System 20, Dec2012, Vol. 20 Issue 6: Special Issue on Computational Definitions of Privacy and Anonymity. {% Seems to discuss HARA utility in detail %} Ingersoll, Jonathan E. Jr. (1987) “Theory of Financial Decision Making.” Rowland and Littlefield, Savage, MD. {% Theoretical analysis of stock market and CAPM under ’92 PT, with effects of probability weighting and loss aversion. %} Ingersoll, Jonathan E. Jr. (2013) “Cumulative Prospect Theory, Aggregation, and Pricing,” working paper. {% %} Ingersoll, Jonathan E. & Stephen A. Ross (1992) “Waiting to Invest: Investment and Uncertainty,” Journal of Business 65, 1–29. {% foundations on statistics; Points out that Fisher did not consider significance levels as objective, and that Pearson was also open to the interpretation of probability as degree of belief. %} Inman, Henry F. (1994) “Karl Pearson and R.A. Fisher on Statistical Tests: A 1935 Exchange from Nature,” American Statistician 48, 2–11. {% one-dimensional utility: uses a kind of mixture-continuity, some weaker than Debreu’s (1959) continuity. %} Inoue, Tomoki (2010) “A Utility Representation Theorem with Weaker Continuity Condition,” Journal of Mathematical Economics 46, 122–127. {% P. 1171: N = 122. Do hypothetical choice because of losses in three-color Ellsberg. ambiguity seeking for losses: they find ambiguity aversion for losses. However, as usual in this case, they did not control for suspicion (suspicion under ambiguity). Subjects could not choose the color to gamble on. What the authors call subadditivity is the usual violation of the s.th.pr. in Ellsberg 3-color. They do not do neuromeasurement but cite much literature on it. %} Inukai, Keigo & Taiki Takahasi (2009) “Decision under Ambiguity: Effects of Sign and Magnitude,” International Journal of Neuroscience 119, 1170–1178. {% DOI: HTTP://DX.DOI.ORG/10.1371/journal.pmed.0020124 %} Ioannidis, John P. A. (2005) “Why Most Published Research Findings Are False,” PLoS Medicine. {% Telling patient that an elective 1-hour procedure has 0.01% probability of death may be hard for people to relate to. Comparing to similar risks, such as same-age and same-sex people having a 0.01% death risk over 1 month, may help. This paper proposes several such ways to explain. Reminds me of an idea of Ron Howard (1988), to introduce a new unit for a small probability of dying, the microort, to help people in communication. %} Ioannidis, John P. A. (2013) “Expressing Death Risk as Condensed Life Experience and Death Intensity,” Medical Decision Making 33, 860–868. {% DOI 10.1007/s11166-016-9245-8 gender differences in risk attitude: women are more risk averse than men. They measure risk aversion assuming EU and finding CRRA. They measure subjective discount rate by fitting hyperbolic discounting, where they take some indifferences and assume linear utility. Because they have utility curvature for CRRA risk aversion they could use this utility function to correct discounting for utility curvature, as in Andersen et al. (2008 Econometrica) and others. But they are not clear on whether they did so and probably they didn't, and simply assumed linear utility. The latter is better than the Andersen et al. method because EU utility is more distorted by nonEU risk factors than that it brings true utility for risk, let be for intertemporal. For risk and time attitudes, they consider two different outcomes: money and number of plants planted that are good for the environment. They call these monetary and environmental environments. The differences they claim in risk and time attitudes can be due simply to different utility of the two kinds of outcomes. Utility curvature of money can be different than of plants, as these can be different than for apples, pears, quantity of wine drunk, and so on. %} Ioannou, Christos A. & Jana Sadeh (2016) “Time Preferences and Risk Aversion: Tests on Domain Differences,” Journal of Risk and Uncertainty 53, 29–54. {% %} Irtel, Hans (1987) “A Conjoint Grassmann Structure for Testing the Additivity of Binocular Color Mixtures,” Journal of Mathematical Psychology 31, 192–202. {% Seems to describe optimism. %} Irwin, Francis W. (1953) “Stated Expectations as Functions of Probability and Desirability of Outcomes,” Journal of Personality 21, 329–335. {% real incentives/hypothetical choice: small differences/same effects %} Irwin, Julie R., Gary H. McClelland, & William D. Schulze (1992) “Hypothetical and Real Consequences in Experimental Auctions for Insurance against Low-Probability Risks,” Journal of Behavioral Decision Making 5, 107–116. {% Seem to find even negative correlation between risk aversion measurements in different contexts. %} Isaac, R. Mark & Duncan James (2000) “Just Who Are You Calling Risk Averse,” Journal of Risk and Uncertainty 20, 177–187. {% Finds that people are more risk averse if they feel good. %} Isen, Alice M. (1993) “Positive Affect and Decision Making.” In Michael Lewis & Jeanette M. Haviland-Jones (eds.) Handbook of Emotions, 261–277, Guilford Press, New York. {% Proposes a bad-deal aversion to explain data better than with loss aversion. %} Isoni, Andrea (2011) “The Willingness-to-Accept/Willingness-to-Pay Disparity in Repeated Markets: Loss Aversion or ‘Bad-Deal’ Aversion?,” Theory and Decision 71, 409–430. {% Redo Plott & Zeiler (2005), and confirm it for mugs but not at all for lotteries. Thus criticize the generality claims of P&Z, and suggest that P&Z’s nonreporting of their lottery data is unfortunate. In their reply, P&Z explain that their lottery data were only meant for learning, and contained many anomalies making them too unreliable. P&Z disagree with many other things. Oh well, I think that loss aversion is strong but volatile, and small details can change it. %} Isoni, Andrea, Graham Loomes, & Robert Sugden (2011) “The Willingness to PayWillingness to Accept Gap, the “Endowment Effect”, Subject Misconceptions, and Experimental Procedures for Eliciting Valuations: Comment,” American Economic Review 101, 991–1011. {% Nice. %} Ito, Kiyosi (1996, ed.) “Encyclopedic Dictionary of Mathematics; 3rd edn.; translated from Japanese. MIT, Cambridge, MA. {% Public good games with framing as gain or as loss (in latter case subjects first get endowed with the public good). Prospect theory’s predictions work. The paper uses repeated payments in each game again, and not a RIS. This in itself can move in the direction of expected value. %} Iturbe-Ormaetxe, Iñigo, Giovanni Ponti, Josefa Tomás, & Luis Ubeda (2011) “Framing Effects in Public Goods: Prospect Theory and Experimental Evidence,” Games and Economic Behavior 72, 439–447. {% probability elicitation: applied to experimental economics. Measures matching probabilities of the right to play a strategic game against an opponent. Interprets playing the game as ambiguity. This is often done. One usually does not know the probability of what the opponent does. But a difference may be that strategic considerations concern more than what is usually called uncertainty (or ambiguity). Very correctly, points out that we can’t measure ambiguity attitude without speculating on beliefs. However, belief is then simply measured by direct questioning, nonincentivized (discussed in §7.2), and is taken to be additive. This is what some (Fox, Tversky, Wu, Gonzalez) have called the two-stage model, although they allowed for nonadditive beliefs. Next, matching probabilities are measured (if I understand right) from binary choices between playing the game and playing a lottery. From this, subjects are categorized into three categories: ambiguity averse, ambiguity neutral, and ambiguity seeking. They are also divided into three categories of risk averse, risk neutral, or risk seeking, and in three categories regarding sophistication or naivite (naïve is taken here very strictly to mean not reckoning at all with the opponent’s side and taking the probabilities over his strategies uniformly; 10% of the subjects will be that) versus sophisticated (reckoning with other’s plans in any way). The percentages of ambiguity seeking, ambiguity neutrality, and ambiguity averse are 32/46/22, so that ambiguity aversion is the least prevalent. Not very surprising given that here other, strategic, aspects play a role. (game theory as ambiguity) P. 367 4th para rightfully says that nonadditive capacities are too general, and then assumes in fact the source method of Abdellaoui et al. (2011): probabilistic sophistication within the ambiguous (meaning game) source and the risky source, with the weighting function (I would call it source function) different for the two sources so no global probabilistic sophistication. P. 369 para 4 erroneously cites Epstein for this approach. Epstein took probabilistic sophistication as designating unambiguity (risk), and took deviations from probabilistic sophistication as ambiguity. He with much emphasis did not want any exogenous concept of unambiguous. Thus if there is probabilistic sophistication within two sources, as is the case here (and as also in Ellsberg 2-color), then he had no tool for saying which is unambiguous (his event-derivatives are impractical in this experiment, as everywhere). Ivanov takes neo-additive weighting functions with only one parameter, the pessimism parameter, by multiplying beliefs by (1c) (p. 360 para -2). Thus he can only capture the pessimism component, and only the positive part of it (negative pessimism, i.e., optimism, is excluded beforehand) and he also does not capture the orthogonal insensitivity component. correlation risk & ambiguity attitude: the author does not explicitly discuss this, but from Figure 2 (p. 384) lowest panel one can see that risk aversion is negatively correlated with ambiguity aversion, where the latter is described above is not just common ambiguity but also involves preference for strategic uncertainty. %} Ivanov, Asen (2011) “Attitudes to Ambiguity in One-Shot Normal-Form Games: An Experimental Study,” Games and Economic Behavior 71, 366–394. {% Seem to use sophistocated probabilistic choice-statistical re-analysis of Tversky (1969, Intransitivity of Preferences) that casts doubt on whether there really was intransitivity in the data. %} Iverson, Geoffrey I. & Jean-Claude Falmagne (1985) “Statistical Issues in Measurement,” Mathematical Social Sciences 10, 131–153. {% A remarkable variation of the smooth KMM model. For the 2nd order acts the authors do not impose EU axioms, but Yaari’s (1987) dual axioms (which means giving up smoothness). Linear utility in the 2nd stage is very reasonable because 1st stage utils are input here. They are kind of axiomatizing using RDU for ambiguity! %} Iwaki, Hideki & Yusuke Osaki (2014) “The Dual Theory of the Smooth Ambiguity Model,” Economic Theory 56, 275–289. {% %} Iyanaga, Shôkichi & Yukiyosi Kawada (1977/1980) “Encyclopedic Dictionary of Mathematics, Vols 1 and 2.” Mit-Press, Cambridge, MA. {% One should watch out that probability can mean nonadditive measure in this paper (footnote 4). But the probability measures in the prior set P, and the second-order measure , are meant to be additive (Izhakian, personal communication, April 24, 2017). The author introduces a new ambiguity model (name: see title of paper), mixing ideas of the smooth model with Schmeidler’s RDU. It takes a two-stage approach as the smooth model does. For risk, known probabilities, it still assumes EU so that a vNM utility function U captures risk attitude. But then, unlike smooth, the second order integral does not involve an extra utility transformation, but an RDU Choquet-type integral with the nonadditive measure capturing ambiguity attitude. Whereas an ambiguity-neutral person would use goodnews probabilities that are linearly (through 2nd order probabilities) weighted averages of goodnews probabilities, the author here inserts a transformation  (this is a capital upsilon and it is called the outlook function) on [0,1] giving a quasilinear mean, doing mathematically with goodnews probabilities what certainty equivalents in EU do with outcomes. Concavity of the transformation pushes down all the resulting goodnews probabilities, bringing extra pessimism and, hence, ambiguity aversion, and with convexity it all is opposite. Interestingly, if the transformation is S-shaped (opposite of inverse-S) in the sense of convex then concave then this gives likelihood insensitivity: the goodnews probabilities for best outcomes are small and all move in the area where the transformation is convex, giving overestimation and extra optimism there. For worst outcomes we similarly get underestimation of the goodnews probabilities and extra pessimism. Because these things do not involve direct convex combinations I can’t see through the behavioral implications completely. Also it seems that not the absolute level of the transformation , but its local degree of convexity/concavity, determines its effects here. Besides the outlook function there also is a capital gamma function that further affects how the events are weighted in an overall Choquet-type integral.
The model has attitudes referring to (the set of) probabilities and, thus, is event-driven rather than outcome-driven. (event/utility driven ambiguity model: event-driven). For losses, the author does not use a reflected integral, as with PT, but the same integral, as with CEU/RDU. If I understand right, this is taken as reference- or sign dependence. For the aversion to mean-preserving spreads to which concavity of is equivalent, we need 2nd order probabilities exogenously given to make this directly observable. %}
Izhakian, Yehuda (2017) “Expected Utility with Uncertain Probabilities Theory,” Journal of Mathematical Economics 69, 91–103. {% %} Izhakian, Yehuda (2016) “A Theoretical Foundation of Ambiguity Measurement,” working paper. {% Calculates uncertainty premium for smooth model in money units. %} Izhakian, Yehuda & Menachem Brenner (2011) “The Uncertainty Premium in an Ambiguous Economy,” The Quarterly Journal of Finance 1, 323–354. {% loss aversion: erroneously thinking it is reflection: P. 68, footnote 10: “we do not assume different attitudes toward risk for losses and for gains (i.e., loss aversion).” This paper uses Izhakian’s ambiguity model to fit data from the financial market. It takes an, exogenously set, two-stage model of ambiguity like the smooth model only using RDU-type goodnews probability transformation rather than the utility-transformation of the smooth model. Then it does parametric data fitting to assess 2nd order beliefs and the rest. It finds that, whereas extra risk-volatility makes people exercise options later, a common and plausible phenomenon, ambiguity does the opposite. %} Izhakian, Yehuda & David Yermackyz (2017) “Risk, Ambiguity, and the Exercise of Employee Stock Options,” Journal of Financial Economics 124, 65–85. {% Z&Z %} Jack, William & Louise Sheiner (1997) “Welfare-Improving Health Expenditure Subsidies,” American Economic Review 87, 206–221. {% %} Jackson, Mathew O. (1986) “Continuous Utility Functions in Consumer Theory (A set of duality theorems),” Journal of Mathematical Economics 15, 63–77. {% Course, not survey, on implementation. It does not take game theory as given, but investigates what game structure has best properties, which makes it primarily a normative field. §7.1, p. 691 ff., discusses incentive compatibility, which becomes an issue in games with incomplete and private information. %} Jackson, Mathew O. (2001) “A Crash Course in Implementation Theory,” Social Choice and Welfare 18, 655–708. {% This paper starts from the well-known fact that time inconsistency at group level can be generated from aggregation where all individuals are time consistent. It experimentally tests it. 3/4 of subjects is present-biased and 1/4 future-biased or unspecified. So as to separate genuine time preference (as of consumption) from market-driven cash-flow, they use a special system of paying in tokens leading to discounted payoffs (p. 4192 bottom). %} Jackson, Matthew O. & Leeat Yariv (2014) “Present Bias and Collective Dynamic Choice in the Lab,” American Economic Review 104, 4184–4204. {% Shows further problems of aggregating time preferences under heterogeneity: any Pareto optimal nondictatorial rule must be time inconsistent. They also obtain intransitivity results. %} Jackson, Matthew O. & Leeat Yariv (2015) “Collective Dynamic Choice: The Necessity of Time Inconsistency,” American Economic Journal: Microeconomics 7, 150–178. {% %} Jackwerth, Jens Carsten & Mark Rubinstein (1996) “Recovering Probability Distributions from Option Prices,” Journal of Finance 51, 1611–1631. {% Newcomb’s paradox %} Jacobi, Northon (1993) “Newcomb’s Paradox: A Realistic Resolution,” Theory and Decision 35, 1–17. {% ISBN 0521635381) %} Jacobs, Donald P., Ehud Kalai, & Morton I. Kamien (1998) “Frontiers of Research in Economic Theory: The Nancy L. Schwartz Memorial Lectures, 1983 - 1997.” Cambridge University Press. {% Children use more base-rates as they get older; use of representativeness heuristic is also examined. %} Jacobs, Janis E. & Maria Potenza (1991) “The Use of Judgment Heuristics to Make Social and Object Decisions: A Development Perspective,” Child Development 62, 166–178. {% Nash equilibrium discussion %} Jacobsen, Hans-Jörgen (1996) “On the Foundations of Nash Equilibrium,” Economics and Philosophy 12, 67–88. {% law and decision theory: subject had to predict jury decisions, receiving info on judgments by others. Decision bias (discount information of others too much in our decisions) in a law context. Stronger with real experienced attorneys (although better calibrated) than with students. Experience enhances costly mistake! %} Jacobson, Jonas, Jasmine Dobbs-Marsh, Varda Liberman, & Julia A. Minson (2011) “Predicting Civil Jury Verdicts: How Attorneys Use (and Misuse) a Second Opinion,” Journal of Empirical Legal Studies 8, 99–119. {% %} Jacobson, Sarah & Ragan Petrie (2009) “Learning from mistakes: What Do Inconsistent Choices over Risk Tell Us?,” Journal of Risk and Uncertainty 38, 143–158. {% %} Jacowitz, Karen E., and Daniel Kahneman (1995) “Measures of Anchoring in Estimation Tasks,” Personality and Social Psychology Bulletin 21, 1161–1166. {% equity-versus-efficiency: %} Jacquement, Nicolas & Adam Zylbersztejn (2014) “What Drives Failure to Maximize Payoffs in the Lab? A Test for the Inequality Aversion Hypothesis,” Review of Economic Design 18, 243–264. {% %} Jaffray, Jean-Yves (1974) “Existence, Propriétés de Continuité, Additivité de Fonctions d’Utilité sur un Espace Partiellement ou Totalement Ordonné.” Ph.D. dissertation, Université de Paris VI, Paris. {% cancellation axioms: Necessary and sufficient conditions for additive representability in full generality. %} Jaffray, Jean-Yves (1974) “On the Extension of Additive Utilities to Infinite Sets,” Journal of Mathematical Psychology 11, 431–452. {% one-dimensional utility %} Jaffray, Jean-Yves (1975) “Existence of a Continuous Utility Function: An Elementary Proof,” Econometrica 43, 981–983. {% %} Jaffray, Jean-Yves (1975) “Semicontinuous Extension of a Partial Order,” Journal of Mathematical Economics 2, 395–406. {% EU+a*sup+b*inf %} Jaffray, Jean-Yves (1988) “Choice under Risk and the Security Factor,” Theory and Decision 24, 169–200. {% %} Jaffray, Jean-Yves (1988) “Applications of Linear Utility Theory to Belief Functions.” In Bernadette Bouchon & Ronald R. Yager (eds.) (eds.) Uncertainty and Intelligent Systems, Springer, Berlin. {% %} Jaffray, Jean-Yves (1989) “Généralisation du Critère de l’Utilité Espérée aux Choix dans l’Incertain Régulier,” RAIRO-RO/Operations Research 23, 237–267. {% event/utility driven ambiguity model: partly event-driven, through belief-function limits of contained objective-probability events, but also partly outcome-driven, through the function that depends on the minimal outcome m and the maximal outcome M. Gives a separation of ambiguity in decision situation and ambiguity attitude. Let X be an outcome set (it is that more than a state space, as it is originally called in this paper), F the set of all belief functions (could be extended to capacities) on X, and let a preference relation over F be given. We can mix belief functions, and impose the usual vNM mixture-independence condition on preferences (this is best conceivable if the belief functions are exogenously given). It characterizes a preference functional over belief functions, linear w.r.t. mixing. Through Möbius inverses, belief functions can be considered linear mixtures of elementary set functions eB (eB(A) is 1 if A contains B, and is zero otherwise). Under a monotonicity axiom, amounting to complete absence of information for such elementary set functions, the preference value of B can depend only on its supremum and infimum (like -Hurwicz but depends on outcomes). This can be taken as ambiguous outcomes vs. ambiguous probabilities, but properly assumed to concern the state space. We can interpret the mixture weights of the Möbius inverse as probabilized uncertainty, and the eBs as the nonprobabilized information which is to be treated as total absence of information, so as to avoid any subjective inputs (this latter avoidance is a central point in all of Jaffray's work). A justification of mixture operation for belief functions can be found in Jaffray’s 1991 publication in the FUR-IV proceedings (Chikan ed.). %} Jaffray, Jean-Yves (1989) “Linear Utility Theory for Belief Functions,” Operations Research Letters 8, 107–112. {% Dutch book: for belief functions, by using the linear structure of belief functions. %} Jaffray, Jean-Yves (1989) “Coherent Bets under Partially Resolving Uncertainty and Belief Functions,” Theory and Decision 26, 99–105. {% updating %} Jaffray, Jean-Yves (1990) “Bayesian Updating and Belief Functions.” Proceedings of the 3rd International Conference on Information Processing and Management of Uncertainty in Knowledge-Based Systems (IPMU’90), Paris, July 1990, 449–451 (published by ENSTA, Paris). {% §2 briefly discusses the separation of ambiguity in decision situation and ambiguity attitude. %} Jaffray, Jean-Yves (1991) “Belief Functions, Convex Capacities, and Decision Making.” In Jean-Paul Doignon & Jean-Claude Falmagne (eds.) Mathematical Psychology: Current Developments, 127–134, Springer, Berlin. {% This paper justifies the independence for belief functions. Link to paper %} Jaffray, Jean-Yves (1991) “Linear Utility Theory and Belief Functions: A Discussion.” In Atilla Chikan (ed.) Progress in Decision, Utility and Risk Theory. Kluwer Academic Publishers, Dordrecht. {% Proposes a way to update belief functions, and proves that this method, unlike the Dempster/Shafer method, will again yield a belief function. One direction of the result was obtained independently by Fagin & Halpern (1991), but this paper very nicely adds the more difficult direction, showing equivalence. %} Jaffray, Jean-Yves (1992) “Bayesian Updating and Belief Functions,” IEEE Transactions on Systems, Man, and Cybernetics 22, 1144–1152. {% event/utility driven ambiguity model: event-driven Download 7.23 Mb. Share with your friends:
|