Advances in Model-Based Decision Support
School of Business | Doctoral thesis (article-based)
Unless otherwise stated, all rights belong to the author. You may download, display and print this publication for Your own personal use. Commercial use is prohibited.
32 + app. 70
Aalto University publication series DOCTORAL THESES, 18/2022
AbstractThis dissertation consists of three individual publications addressing on two important classes of decision analysis problems. Publications I and II contribute to the field of decision making under uncertainty, whereas Publication III contributes to that of multi-criteria decision making (MCDM) in a riskless decision context. Publication I-III contain both original theoretical, methodological contributions and also extensive empirical analyses. Publications I-III together constitute a substantial contribution to decision analytics related literature in Operations Research and Management Science (ORMS). Publication I develops stochastic dominance (SD) criteria under incomplete information on state probabilities. Specifically, we identify portfolios that dominate a given benchmark for any state probabilities in a given set. The proposed approach is applied to analyze if industrial diversification can be utilized to outperform the market portfolio. The results from this application demonstrate that the use of set-valued state probabilities can help to improve out-of-sample performance of SD-based portfolio optimization. Publication II develops novel robust second-order stochastic dominance (SSD) criteria to capture the strength of dominance and portfolio optimization models utilizing these criteria to identify portfolios whose in-sample SSD dominance over a given benchmark is likely to hold also out-of-sample. The developed models can incorporate incomplete probability information by allowing a set of feasible state probabilities. We also show that these portfolio optimization models can be formulated as linear programming problems. We report results from applying these SSD-based portfolio optimization models with different sets of state probabilities in an empirical application, with a focus on evaluating the out-of-sample portfolio performance of the optimized portfolios. Publication III compares and contrasts two types of choice strategies in a riskless, multi-objective context. Specifically, in the win-win strategy, the developed choice model is based on standard theoretical assumptions from microeconomic theory that the underlying marginal value functions are increasing and concave. Additionally, the aggregate value function is assumed additive and separable, and decision maker knows which good brings most value at the moment of choice, if added to the basket. On the other hand, the trade-off strategy is based on Tversky-Kahneman's reference-dependent model with loss aversion in riskless choice, where the value functions are concave for gains and convex for losses.
Supervising professorLiesiö, Juuso, Prof., Aalto University, Department of Information and Service Management, Finland
Thesis advisorLiesiö, Juuso, Prof., Aalto University, Department of Information and Service Management, Finland
data-driven analytics, decision support models, risk and uncertainty, stochastic dominance, portfolio optimization, robustness, incomplete information, MCDM, choice behavior
[Publication 1]: Juuso Liesiö, Peng Xu, and Timo Kuosmanen. Portfolio diversiﬁcation based on stochastic dominance under incomplete probability information. European Journal of Operational Research, 286, 2, 755-768, Oct 16 2020.
Full text in Acris/Aaltodoc: http://urn.fi/URN:NBN:fi:aalto-202004032710DOI: 10.1016/j.ejor.2020.03.042 View at publisher
- [Publication 2]: Peng Xu. New approaches for identifying robust dominating portfolios based on second-order stochastic dominance. Revised and resubmitted to OR Spectrum, Jan 18 2022
[Publication 3]: Pekka Korhonen, Jyrki Wallenius, Tolga Genc, and Peng Xu.On rational behavior in multi-attribute riskless choice. European Journal of Operational Research, 288, 1, 331-342, Jan 1 2021.
DOI: 10.1016/j.ejor.2020.05.056 View at publisher