Model uncertainty : the effect on robustness, estimation and stochastic optimisation

A.G. Balter

Research output: ThesisDoctoral ThesisInternal

743 Downloads (Pure)

Abstract

Mathematical models form the basis of decisions made in the fields of economics, finance, medicine and climatology. Given that things tend not to follow a mathematical model in real life, there's almost always a mismatch between model-based decisions and their desired effect. This dissertation examines alternative models that are statistically plausible. It also seeks to find an investment strategy that is less sensitive to deviations from the current model. With life expectancy continuing to rise, model-based interest rates will have a considerable impact on the policies developed by pension funds and insurance companies over the next century or so. The amount of interest determines how much money should be set aside to cover future pay-outs. Given the current pension debate, this dissertation identifies the uncertainties associated with the interest term structure.
Original languageEnglish
QualificationDoctor of Philosophy
Awarding Institution
  • Maastricht University
Supervisors/Advisors
  • Pelsser, Antoon, Supervisor
  • Schotman, Peter, Supervisor
Award date19 Feb 2016
Place of PublicationMaastricht
Publisher
Print ISBNs9789461595171
Publication statusPublished - 2016

Keywords

  • uncertainty model
  • statistics
  • mathematical models
  • interest
  • robustness
  • policy
  • investment strategies
  • value engineering

Cite this

Balter, A. G. (2016). Model uncertainty : the effect on robustness, estimation and stochastic optimisation. Datawyse / Universitaire Pers Maastricht.