Bibliography
§4.5 points out that there is little field data validation of Amos
Download 7.23 Mb.
|
- Navigate this page:
- KirstenI
- Principle of Complete Ignorance
- PT, applications
| §4.5 points out that there is little field data validation of Amos’ ideas, but cites some.
§4.6 explains that Amos did not, or little, commit to normative viewpoints. %} Laibson, David I. & Richard J. Zeckhauser (1998) “Amos Tversky and the Ascent of Behavioral Economics,” Journal of Risk and Uncertainty 16, 7–47. {% %} Lajeri, Fatma & Lars Tyge Nielsen (2000) “Parametric Characterizations of Risk Aversion and Prudence,” Economic Theory 15, 469–476. {% %} Lakatos, Imre (1970) “Falsification and the Methodology of Scientific Research Programmes.” In Imre Lakatos & Alan Musgrave (eds.) Criticism and the Growth of Knowledge, Cambridge University Press, Cambridge. {% adaptive utility elicitation: adaptive SG %} Lalonde, Lyne, Ann E. Clarke, Lawrence Joseph, Stephen A. Grover, & Canadian Collaborative Cardiac Assessment Group (1999) “Conventional and Chained Standard Gamble in the Assessment of Coronary Heart Disease Prevention and Treatment,” Medical Decision Making 19, 149–156. {% %} Lamers, Leida M., Peep F.M. Stalmeier, Paul F.M. Krabbe, & Jan J.V. Busschbach (2006) “Inconsistencies in TTO and VAS Values for EQ-5D Health States,” Medical Decision Making 26, 173–181. {% Seem to propose what in fact amounts to Chew’s weighted utility formula for decision under risk. That is, the average of a set of stimuli x1, ..., xn is sum w(xj)s(xj) /sum w(xj) for functions w and s. P. 1 second column next-to-last paragraph seems to describe a nonnormalized sum w(xj)s(xj). P. 5 second column third paragraph last sentence also suggests it a bit. There are no formulas and it is nowhere very clear. %} Lampel, Anita K. & Norman H. Anderson (1968) “Combining Visual and Verbal Information in an Impression-Formation Task,” Journal of Personality and Social Psychology 9, 1–6. {% ambiguity seeking for losses: do not find that, but find ambiguity aversion there, in Ellsberg choices. They, however, do not control for suspicion (suspicion under ambiguity). It is like Smith, Dickhaut, McCabe, & Pardo (2002) who also find ambiguity aversion for losses but, also, do not control for suspicion. %} Lan, Cherng-Horng, Peter Ayton, & Nigel Harvey (2010) “The Role of Decision Consequences in Ambiguity Aversion,” working paper. {% Poster presented at SPUDM Stockholm 2005. They present the classical Ellsberg paradox, but frame the options in a matrix with four states of nature (Red known & Black unknown, etc.) as four columns in a matrix. (ambiguity seeking) In this format, the paradox disappears and people are ambiguity neutral! %} Lan, Cherng-Horng & Nigel Harvey (2005) “How Would Savage Frame Ellsberg’s Two-Color Problem?,” {% Kirsten&I; Characterizes something like discounted utility for continuous time but, like Fishburn & Rubinstein (1982), with only one consumption at one time point. DC = stationarity; Axiom P6 is nice version of dynamic consistency, and axiom P. 10 a nice version of stationarity. %} Lancaster, Kelvin J. (1963) “An Axiomatic Theory of Consumer Time Preference,” International Economic Review 4, 221–231. {% Recommended to be by Hahneman in Aug. 2000. The paper adds an extra layer to commodities, something like features of those. (For example, my example: not commodities are what it is really about, but the shelter or health improvement that they give us.) Gives background to complementarity/substitutability. %} Lancaster, Kelvin J. (1966) “A New Approach to Consumer Theory,” Journal of Political Economy 74, 132–157. {% paternalism/Humean-view-of-preference: gives long list of reasons for not deleting responses deemed irrational (and not one reason for deleting them). They can be summarized as: it is wrong for those responses that are not irrational so that they were misdeemed. It is like writing a long list of reasons for why a null hypothesis can be rejected incorrectly, ending up with the recommendation to never reject a null hypothesis. The authors ascribe empirical meaning to continuity, and claim that most modern research is on preferences and that preferences is not choice but introspection (so, contrary to most, they do not equate preference with binary choice in most of their text). Sometimes seem to follow the unfortunate convention of equation rationality with transitivity and completeness, an unfortunate convention common in revealed preference theory. Give recommendations such as “As a general guide, researchers should consider carefully how they design DCEs [discrete choice experiments]. (p. 807 bottom) and “one should design the largest design possible … given constraints such as research budgets as well as more subjective constraints regarding number of attributes and complexity” (p. 808 top). P. 799 qualifies a self-reference as “pioneering.” %} Lancsar, Emily & Jordan Louviere (2006) “Deleting `Irrational’ Responses from Discrete Choice Experiments: A Case of Investigating or Imposing Preferences?,” Health Economics 15, 797–811. {% Seems that they propose 0.61 as threshold for substantial correlation. %} Landis, J. Richard & Gary G. Koch (1977): “The Measurement of Observer Agreement for Categorical Data,” Biometrics 33, 159–174. {% Principle of Complete Ignorance: axioms for preferences over intervals, interpretable as complete ignorance. %} Landes, Jürgen (2014) “Min–Max Decision Rules for Choice under Complete Uncertainty: Axiomatic Characterizations for Preferences over Utility Intervals,” International Journal of Approximate Reasoning 55, 1301–1317. {% %} Landsberger, Michael & Isaac Meilijson (1990) “A Tale of Two Tailes: An Alternative Characterization of Comparative Risk,” Journal of Risk and Uncertainty 3, 65–82. {% %} Landsberger, Michael & Isaac Meilijson (1990) “Lotteries, Insurance, and Star-Shaped Utility Functions,” Journal of Economic Theory 52, 1–17. {% %} Landsberger, Michael & Isaac Meilijson (1990) “Demand for Risky Financial Assets: A Portfolio Analysis,” Journal of Economic Theory 50, 204–213. {% foundations of statistics: paper argues that probability is better learned using experiments than using maths. %} Lane, Andrew (2009) “Experimental Probability in Elementary School,” Teaching Statistics 31, 34–36. {% Argues in fact for violation of RCLA! He argues for the following difference. Imagine T is a sufficient statistic. First assume that in a first stage a value t of T is generated. In a second stage, conditional upon that value t of T, a corresponding value x of the observed statistic X is observed corresponding with T (so in T’s inverse of t). Note that the second-stage probability distribution is independent of the parameter . In this two-stage process, Hill finds sufficiency convincing. In general, when the two stages are collapsed together, he does not find it convincing! %} Lane, David A. (1984) “Discussion” in Berger, James O. & Robert L. Wolpert (1984) “The Likelihood Principle: A Review, Generalizations and Statistical Implications.” Lecture Notes, Monograph Series, Volume 6, Institute of Mathematical Statistics, Hayward, California; 2nd edn. 1988; pp. 175–181. {% crowding-out: Ch. 19 seems to survey the crowding out effect as studied by psychologists. %} Lane, Robert E. (1991) “The Market Experience.” Cambridge University Press, New York. {% Lotteries for charitable purposes work better than voluntary gifts; paper pays special attention to risk attitudes of potential donateurs, and the heterogeneity of those risk attitudes, and that this may sometimes imply that multiple-outcome lotteries work better than single-outcome lotteries and have some predictions confirmed in an experiment. They use EU to analyze throughout and do not mention nonEU. %} Lange, Andreas, John A. List, & Michael K. Price (2007) “Using Lotteries to Finance Public Goods: Theory and Experimental Evidence,” International Economic Review 48, 901–927. {% %} Lange, Andreas & Anmol Ratan (2010) “Multi-Dimensional Reference-Dependent Preferences in Sealed-Bid Auctions – How (Most) Laboratory Experiments Differ from the Field,” Games and Economic Behavior 68, 634–645. {% strength-of-preference representation; Points out that comparability of strength of preference determines utility up to level and unit; i.e., utility is “measurable” in the terminology of those days. Refers to Frisch (1926) for a formal analysis. Gives reference to many who overlooked this point. Argues that observable choice gives only ordinal utility and that that is all needed for equilibrium. For strength of preference, psychological introspection is needed. Says that the latter is needed for a theory of “human welfare” but does not explain the latter. %} Lange, Oskar (1934) “The Determinateness of the Utility Functions,” Review of Economic Studies 1, 218–224. {% conservation of influence: through illusion of control. We treat chance events as if they involve skill and therefore as if we have control over them %} Langer, Ellen J. (1975) “The Illusion of Control,” Journal of Personality and Social Psychology 32, 311–328. {% We treat chance events as if they involve skill and therefore as if we have control over them %} Langer, Ellen J. (1977) “The Psychology of Chance,” Journal of Social Theory and Behavior 7, 185–207. {% Usually, separate evaluation of a number of lotteries comes out lower than their joint evaluation (so, of their convolution), because in the second case many losses are neutralized by gains so that the loss aversion effects are less strong. There do exist special lotteries such that the separate evaluation of two of them comes out higher than the joint evaluation. This is pointed out in this paper, and implications are discussed. They suggest (e.g. p. 730 l. 2) that subjects, in complex decisions, may simply go by the probability of attaining some target, e.g. they may minimize the probability of losing. %} Langer, Thomas & Martin Weber (2001) “Prospect Theory, Mental Accounting, and Differences in Aggregated and Segregated Evaluation of Lottery Portfolios,” Management Science 47, 716–733. {% P. 46 suggests a bit that Jevons introduced outcomes in terms of final wealth, and that Bentham had them as changes w.r.t. reference point. The authors use different terminologies than I am used to, and I should not make the mistake of reading modern ideas into old (Bentham) writings, and it isn’t 100%. The authors write that Jevons turned preferences into “exogenous” and unchanging, and that with Bentham it was “endogenous” and “changing” depending on preceding pains, pleasures (and, hence, decisions which explains the endogenekity). In their formal model later they also bring in time explicitly to capture the changes. This “changing” is a broad term that could mean anything. Yet I think that they really mean changing only in the sense of reference-dependence. P. 46: “Consequently, the agent’s preference order will be viewed as depending on his initial situation, and on asymmetric sensitivity to gains and losses, relative to this situation (§2). Bentham clearly expressed this idea when he argued that `the pleasure of gaining is not equal to the evil of losing’ (1785-6: 331).” P. 47-48 acknowledges that there is no direct evidence for “endogenous” (what I call reference dependence) preference in Bentham, but that indirect evidence is conclusive. P. 50 acknowledges the value of the ordinal revolution (this is my interpretation): “founding economic calculus on the basis of a given utility function was already a difficult task, which required nearly a century after Jevons to be achieved; but the enterprise would surely have been bound to fail with a utility function submitted to continuous changes.” P. 52: “it opens the path to the possibility that a same final situation of alternative trajectories is associated with different levels of utility.” Pp. 66-67: “the juncture between the positive and the normative aspects of the principle of utility.” paternalism/Humean-view-of-preference: §5 argues that Bentham advocated paternalism where biases (mistakes in felicific calculus) are to be corrected and reduced. %} Lapidus, André & Nathalie Sigot (2000) “Individual Utility in a Context of Asymmetric Sensitivity to Pleasure and Pain: An Interpretation of Bentham’s Felicific Calculus,” European Journal of the History of Economic Thought 7, 45–78. {% A correction of Zimper (2011). updating for nonadditive measures is discussed. %} Lapied, André & Pascal Toquebeuf (2013) “A Note on “Re-Examining the Law of Iterated Expectations for Choquet Decision Makers,” Theory and Decision 74, 439–445. {% Pr. of insufficient reason; seems to have stated the gambler’s fallacy somewhere (Peter Ayton). Was he the first? p. xvii in reprint in Oeuvres completes de Laplace, Voi. 7, Gauthier-Villars, Paris, 1886 seems to state the rule of succession (name given later by Venn (1888): if on n trials we see m successes, then then next trial has success probability (m+1)/(n+1). (The rule I use privately lifelong.). %} Laplace, Pierre Simon de (1796) “Essai Philosophique sur les Probabilités.” Paris. (5th edn. 1825). Translated into English as “A Philosophical Essay on Probabilities,” Dover Publications, New York, 1951. {% p. 402 in reprint in Oeuvres completes de Laplace, Voi. 7, Gauthier-Villars, Paris, 1886 seems to state the rule of succession (name given later by Venn (1888): if on n trials we see m successes, then then next trial has success probability (m+1)/(n+1). (The rule I use privately lifelong.). %} Laplace, Pierre Simon de (1812) “Théorie Analytique des Probabilités.” Courcier, Paris, 2nd ed., 1814; 3rd ed., 1820. {% Paper says that Bernoulli’s theory and prospect theory (here the paper is just plainly wrong) do not permit individual differences in risk attitude, are called “universal theories” for that reason, and are contrasted with individual-difference theories, to which belong EU, Lopes’ theory, and Atkinsons theory (latter turns out to consider events under control of the participant) %} Larrick, Richard P. (1993) “Motivational Factors in Decision Theories: The Role of Self-Protection,” Psychological Bulletin 113, 440–450. {% Seem to argue that taking averages in expert aggregation is better than often thought. %} Larrick, Richard P. & Jack B. Soll (2006) “Intuitions about Combining Opinions: Misappreciation of the Averaging Principle,” Management Science 52, 111–127. {% Change in Miles-per-Gallon from 12 to 14 has a larger impact on fuel reduction than from 28 to 40. This has a bit to do with well-known mistake to take 1/X linear iso convex in X. For example, driving half the way with speed 100 h and half way with speed 300/h is slower than driving 200/h all the way, but many take it to go equally fast. This is vaguely related to: ratio-difference principle. %} Larrick, Richard P. & Jack B. Soll (2008) “The MPG Illusion,” Science 20, 1592–1594. {% Seem to show that gains and losses are psychologically distinct. %} Larsen, Jeff T., A. Peter McGraw, Barbara A. Mellers, & John T. Cacioppo (2004) “The Agony of Victory and the Thrill of Defeat: Mixed Emotional Reactions to Disappointing Wins and Relieving Losses,” Psychological Science 15, 325–330. {% scheurkalender etc. van Bert en mij %} Larson, Gary {% second-order probabilities to model ambiguity; presented 2nd order probabilities to subjects, with 20 possible compositions of 100 balls, where the 2nd order distribution was too complex to be reduced. Subjects preferred small variance of 2nd order distributions to big variances under same expectation, violating RCLA. %} Larson, James R., Jr. (1980) “Exploring the External Validity of a Subjectively Weighted Utility Model of Decision Making,” Organizational Behavior and Human Performance 26, 293–304. {% The whole issue of the journal is dedicated to infinity. %} Larvor, Brendan P., Benedikt Löwe, & Dirk Schlimm (2015) “History and Philosophy of Infinity,” Synthese 192, 2339–2344. {% %} Laskey, Katheryn B. & Paul E. Lehner (1988) “Belief Maintenance: An Integrated Approach to Uncertainty Management,” Proceedings of the 7th National Conference on AI (AAAI-88) Minneapolis. {% error theory for risky choice %} Laskey, Katheryn B. & Gregory W. Fischer (1987) “Estimating Utility Functions in the Presence of Response Error,” Management Science 33, 965–980. {% intuitive versus analytical decisions; Give alternative explanation for Dijksterhuis et al. (2006) finding. %} Lassiter, G. Daniel, Matthew J. Lindberg, Claudia González-Vallejo, Francis S. Bellezza, & Nathaniel D. Phillips (2009) “The Deliberation-without-Attention Effect: Evidence for an Artifactual Interpretation,” Psychological Science 20, 671–675. {% Footnote 12 says that Bernoulli (1738) is generally credited for being the first to use utility. Argues that maximization of expectation of geometric mean; i.e., Bernoulli’s logarithmic utility, is a useful approach. P. 147 middle of second column points out that the classical expected value criterion left no space for individual variation, so no subjectivity involved. %} Latané, Henry A. (1959) “Criteria for Choice among Risky Ventures,” Journal of Political Economy 67, 144–155. {% inverse-S; use the two-parameter extension of Karmarkar, as Goldstein & Einhorn, 1987) also did, and find inverse-S for both gains and, as it seems, losses. real incentives: they did hypothetical choice %} Lattimore, Pamela M., Joanna R. Baker, & Ann D. Witte (1992) “The Influence of Probability on Risky Choice,” Journal of Economic Behavior and Organization 17, 377–400. {% First paper on program Decision Maker %} Lau, Joseph, Jerome P. Kassirer, & Stephen G. Pauker (1983) “Decision Maker 3.0: Improved Decision Analysis by Personal Computer,” Medical Decision Making 3, 39–43. {% PT, applications, loss aversion; utility concave near ruin & Risk averse for gains, risk seeking for losses: consider losses, and find most risk seeking if no ruin, risk aversion if ruin comes in. %} Laughhunn, Dan J., John W. Payne, & Roy L. Crum (1980) “Managerial Risk Preferences for Below-Target Returns,” Management Science 26, 1238–1249. {% statistics for C/E %} Laupacis, Andreas, David H. Feeny, Alan S. Detsky, & Peter X. Tugwell (1992) “How Attractive Does a New Technology Have to Be to Warrant Adoption and Utilisation? Tentative Guidelines for Using Clinical and Economic Evaluations,” Canadian Medical Association Journal 146, 473–481. {% Participants choose between gaining on the unknown Ellsberg urn (50-50 in normative sense) and gaining with probability p, for varying p. So, this is finding matching probability using choice list for the 50-50 Ellsberg urn. Obviously, the unknown urn is chosen less as p increases. Average switch is before p = .50, in agreement with the commonly found ambiguity aversion for .50-.50 Ellsberg urns. It is also obvious that most preferences will switch around the normative threshold; i.e., around p = .50. Contrary to the authors’ claim, this does not mean that people are more sensitive near .50 than elsewhere in a general sense. reflection at individual level for risk: p. 117: no relation reflection at individual level for ambiguity: p. 117: no relation: “Thus there is sufficient reason to argue that loss trials and gain trials tap different processes.” correlation risk & ambiguity attitude: p. 117: ambiguity aversion is positively related to risk aversion for losses, and is not significantly related to risk attitude for gains. %} Lauriola, Marco & Irwin P. Levin (2001) “Relating Individual Differences in Attitude toward Ambiguity to Risky Choices,” Journal of Behavioral Decision Making 14, 107–122. {% Apparently do only hypothetical choice. Ambiguous urn always is 2-color, but they also vary the total nr. of balls in the urn, and find that this does something even if normatively it shouldn’t. Measure matching probabilities. Claim as novelty that they derive it from bisection, rather than from matching as did Kahn & Sarin (1988). correlation risk & ambiguity attitude: find positive relation between ambiguity attitude and risk aversion. Do so by first experiment to measure ambiguity aversion, then taking the very extremely ambiguity averse and the very extremely ambiguity seeking separately (extreme-group design), and comparing their risk attitudes to find significant differences in the latter. This method does not show much of how strong the attitudes are related, only that they are. The second measurement was deliberately done two months later only. The intermediates are control group. The potential selection bias and nonrepresentativeness is discussed on p. 132 middle of 2nd column, referring to social psychology for this technique. %} Lauriola, Marco, Irwin P. Levin, & Stephanie S. Hart (2007) “Common and Distinct Factors in Decision Making under Ambiguity and Risk: A Psychometric Study of Individual Differences,” Organizational Behavior and Human Decision Processes 104, 130–149. {% random incentive system. Points out that she does not test the isolation effect because no single-choice situation is involved. She tests a Davis & Holt (1993) conjecture (see there). Treatment 1: pay one randomly selected choice from 10 choices made (the random incentive system) Treatment 2: pay all 10 choices made. Treatment 3: pay one randomly selected choice from 10 choices made (the random incentive system but with payments increased). Treatments 1 and 2 give the same result, suggesting no income effect here. Treatments 1 and 3 give different results, with treatment 3 more risk aversion. I think that this finding entails that no income effect occurred, and (decreasing ARA/increasing RRA) that there was increasing RRA. It does not directly test the Davis-Holt conjecture because for that it should have scaled the payments down and not up. %} Laury, Susan K. (2005) “Pay One or Pay All: Random Selection of One Choice for Payment.” {% real incentives/hypothetical choice: use high real incentives ($100 etc.) for some of the subjects (all students). losses from prior endowment mechanism: they do this. For the high payments, they first let subject do another game theory experiment where they made very much money. equate risk aversion with concave utility under nonEU: p. 406: very unfortunately, the authors do not call concave utility what it is (concave utility), but what it is not: risk aversion. The usual concept of risk aversion (preference for EV over prospect) apparently is also called risk aversion. concave utility for gains, convex utility for losses: find it for hypothetical choice. For real choice they rather find risk aversion and concave utility for both gains and losses. reflection at individual level for risk: p. 419, for hypothetical low outcomes finds reflection, with risk aversion (in their terminology) for gains usually going together with risk seeking for losses and risk seeking for gains mostly going together with risk aversion for losses. For real incentives, however, it is very opposite. Risk aversion for gains has majority risk aversion for losses, and risk seeking for gains has majority risk seeking for losses. P. 422: for hypothetical high payment and, even more for real high payment, there is also violation of reflection at the individual level. The econometric analysis later gives no results at the individual level. An attempt to defend reflection against the finding of this paper can be that when implementing losses from prior endowment mechanism, subjects integrate the payments especially if they are high. From that perspective, I could hope to convince the authors to change their conclusion into: for losses better do hypothetical? () %} Laury, Susan K. & Charles A. Holt (2008) “Further Reflections on Prospect Theory.” In Cox, James C. & Glenn Harrison (eds.) Risk Aversion in Experiments, (Experimental Economics, Volume 12) 405–440, JAI Press, Greenwich, CT. {% real incentives/hypothetical choice: for time preferences; This paper pays subjects in probability of gaining a prize. The authors assume EU and then (well, + backward induction) this amounts to linear (risky!) utility, as pointed out by Roth & Malouf (1979), Cedric Smith (1961), and many others. They assume (implicitly, as did Andersen et al. 2008), that EU utility for risk also is utility for intertemporal discounting, and then use this to estimate discounting while reckoning with that utility curvature. Download 7.23 Mb. Share with your friends:
|