Not logged in.
Quick Search - Contribution
Contribution Details
| Type | Master's Thesis |
| Scope | Discipline-based scholarship |
| Title | Testing the Missing-Link Model |
| Organization Unit | |
| Authors |
|
| Supervisors |
|
| Language |
|
| Institution | University of Zurich |
| Faculty | Faculty of Economics, Business Administration and Information Technology |
| Number of Pages | 40 |
| Date | 2016 |
| Abstract Text | A vast amount of experimental research has exposed behavioural regularities in choices under risk and over time that contravene the axioms of the expected utility model and the discounting utility model (Allais, 1953; Kahneman and Tvesky, 1979; Thaler, 1981; Benzion, Rapoport and Yagil, 1989). It is common that individuals tend to distort probabilities and exhibit decreasing discounting rates over time, known respectively as non-linear probability weighting and hyperbolic discounting. There is currently a growing new trend of theoretical research attempting to explain the amounting evidence pointing to the interaction between behaviour under risk and over time. A promising addition to this new line is the model developed by Epper and Fehr-Duda (2012), which is capable of explaining all the interactive manifestations of both behavioural dimensions. Their approach - denoted as the missing-link model - is based upon two key assumptions: inherent uncertainty about the future and non-linear probability weighting, satisfying the critical property of subproportionality. Certainly, the future is uncertain since there is always the possibility that during waiting something wrong occurs and consequently delayed rewards cannot be received. Moreover, the second key assumption is concerned about the perception of such uncertainty. People usually tend to distort probabilities in a way that may cause a preference reversal when probabilities are scaled down, a trait also known as the common ratio effect. Such a behavioural feature can be accommodated by a probability weighting function that satisfies the condition of subproportionality. From a perceptional perspective, this condition implies that the scaling down of the probabilities makes them less differentiable. The missing-link model not only predicts that the presence of inherent uncertainty discount rates increases the level of discount rates just as the standard approaches, as well as predicting that discount rates decrease with time, even if the underlying true rate of time preference is constant. In addition, the model foresees that an increase in inherent uncertainty makes the feature of decreasing discount rates stronger. The purpose of this research is to test the model, focusing on its prediction that greater inherent uncertainty generates more strongly declining discount rates. For this purpose, a monetary incentivised experiment is conducted with ETH and UZH students, where one group of subjects underwent a writing task that stimulates the subjective perception of the inherent uncertainty about the future, while the other group underwent a similar but neutral writing task. Accordingly, one group is primed to more strongly perceive inherent uncertainty before discount rates were elicited. This procedure allows comparing two groups - a priming group and non-priming group - assuming that the priming treatment does not systematically correlate with any personal characteristic that can also affect intertemporal decisions. Overall, the data of 95 subjects formed the basis of the statistical analysis, comprising 51 subjects in the priming group and 44 subjects in the non-priming group. The results reveal some evidence that priming has increased the overall level of discount rates, although no significant evidence was found indicating that priming has increased the degree of decreasing discount rates. It is possible that number of participants per treatment was too low to properly capture group differences in decreasing discount rates. |
| Export |
BibTeX
|