Cookies?
Library Header Image
LSE Theses Online London School of Economics web site

Beyond lucky: measuring and modelling the impact of ‘probability control’ on risky choice

Agarwal, Shweta (2014) Beyond lucky: measuring and modelling the impact of ‘probability control’ on risky choice. PhD thesis, The London School of Economics and Political Science (LSE).

[img]
Preview
PDF - Submitted Version
Download (5MB) | Preview

Abstract

Managers frequently deal with risk by considering uncertainty as an element of the decision problem over which they can exert control — for example, lobbyists trying to exert influence over regulators or managers trying to mitigate Operational Risks related to human processes. This perspective that the probabilities of uncertain events are at times ‘mutable’ — i.e. subject to one’s influence — has an important and previously under-appreciated role in decision-making under risk. The present research, structured as a series of three papers, addresses this gap between theory and practice on the topic of ‘control’ from a descriptive, theoretical and prescriptive perspective. The descriptive paper discusses a novel empirical test of the behavioural effect of ‘control’ on risk taking. The key finding that control does not always enhance risk taking but, instead, has a moderating effect on attitudes to risk, extends insights from related research. Strong preference for exerting control to eliminate uncertainty is also revealed. Affective and cognitive interpretations of the findings are offered and their correspondence with managerial attitudes to risk taking is discussed. The theoretical paper builds on methods in Decision Analysis and Philosophy, and develops a new probability revision rule for modelling control as interventions on uncertainties. This rule is shown to dramatically alleviate the judgmental burden of analysing multiple interventions. Foundational properties for probability revision rules for interventions, similar to the coherence criterion for Bayes rule, are also constructed and a proof that the proposed rule satisfies these properties is offered. In the prescriptive paper, a real world application of the probability revision rule is illustrated in the context of Operational Risk assessment, where several uncertainties are controllable (e.g. staff strikes). It is shown how this rule can be integrated with Operational Risk calculations to explicitly incorporate the effect of managerial mitigations on loss events, thus making a useful contribution to the field. In summary, this research explores the concept of ‘probability control’ as a way to manage risks in the context of Decision Sciences. It furthers our behavioural understanding of risk attitudes to better resonate with managerial perspectives on risk taking and extends the relevance of Decision Analysis methods to corporate risk management.

Item Type: Thesis (PhD)
Additional Information: © 2014 Shweta Agarwal
Library of Congress subject classification: H Social Sciences > HD Industries. Land use. Labor > HD28 Management. Industrial Management
Sets: Departments > Management
Supervisor: Montibeller, Gilberto and Morton, Alec
URI: http://etheses.lse.ac.uk/id/eprint/1021

Actions (login required)

Record administration - authorised staff only Record administration - authorised staff only

Downloads

Downloads per month over past year

View more statistics