Linearly constrained robust portfolio construction using Bauer-Householder bounds on the condition number

dc.contributor.advisorGebbie, Timothy
dc.contributor.authorDalmeyer
dc.contributor.authorLara
dc.date.accessioned2026-09-29T11:39:52Z
dc.date.available2026-09-29T11:39:52Z
dc.date.issued2026
dc.date.updated2026-09-29T11:11:33Z
dc.description.abstractWe investigate and extend the result that the angle between the direction of expected returns and the optimal portfolio weights from unconstrained quadratic portfolio optimisations have an upper bound dependent on the condition number of the covariance matrix. This is known to imply that better conditioned covariance matrices produce weights from unconstrained mean-variance optimisations that are better aligned with each assets expected return. Here we relate the inequality between the angle spanning the expected returns vector α and the optimal weight vector θ – the so called α-weight angle and the condition number, to extend the result to include portfolio optimisations with gearing constraints to provide an extended family of robust optimisations, using the Bauer and Householder inequality. The Bauer and Householder inequality is an extension of the Kantorovich inequality which has been used to establish the unconstrained α-weight angle bounds. We use this to argue that the equally weighted portfolio is in general not preferable to the mean-variance portfolio even with poor forecast ability and a badly conditioned covariance matrix. A new shrinkage methodology directly using the α-weight angle is proposed and investigated, and shown to be superior to a collection of standard shrinkage methodologies known in the literature. We consider simulations for the behavior of the α-weight angle to demonstrate the theoretical results. We argue that even if there is naive knowledge about future expected returns, the new method is preferred.
dc.identifier.apacitation, & (2026). <i>Linearly constrained robust portfolio construction using Bauer-Householder bounds on the condition number</i>. (). University of Cape Town ,Faculty of Science ,Department of Statistical Sciences. Retrieved from http://hdl.handle.net/11427/43840en_ZA
dc.identifier.chicagocitation, and . <i>"Linearly constrained robust portfolio construction using Bauer-Householder bounds on the condition number."</i> ., University of Cape Town ,Faculty of Science ,Department of Statistical Sciences, 2026. http://hdl.handle.net/11427/43840en_ZA
dc.identifier.citation & 2026. Linearly constrained robust portfolio construction using Bauer-Householder bounds on the condition number. . University of Cape Town ,Faculty of Science ,Department of Statistical Sciences. http://hdl.handle.net/11427/43840en_ZA
dc.identifier.ris TY - Thesis / Dissertation AU - Dalmeyer AU - Lara AB - We investigate and extend the result that the angle between the direction of expected returns and the optimal portfolio weights from unconstrained quadratic portfolio optimisations have an upper bound dependent on the condition number of the covariance matrix. This is known to imply that better conditioned covariance matrices produce weights from unconstrained mean-variance optimisations that are better aligned with each assets expected return. Here we relate the inequality between the angle spanning the expected returns vector α and the optimal weight vector θ – the so called α-weight angle and the condition number, to extend the result to include portfolio optimisations with gearing constraints to provide an extended family of robust optimisations, using the Bauer and Householder inequality. The Bauer and Householder inequality is an extension of the Kantorovich inequality which has been used to establish the unconstrained α-weight angle bounds. We use this to argue that the equally weighted portfolio is in general not preferable to the mean-variance portfolio even with poor forecast ability and a badly conditioned covariance matrix. A new shrinkage methodology directly using the α-weight angle is proposed and investigated, and shown to be superior to a collection of standard shrinkage methodologies known in the literature. We consider simulations for the behavior of the α-weight angle to demonstrate the theoretical results. We argue that even if there is naive knowledge about future expected returns, the new method is preferred. DA - 2026 DB - OpenUCT DP - University of Cape Town KW - Mean-variance optimization KW - Bauer and Householder inequality KW - Kantorovich inequality KW - gearing constraints KW - robust optimization LK - https://open.uct.ac.za PB - University of Cape Town PY - 2026 T1 - Linearly constrained robust portfolio construction using Bauer-Householder bounds on the condition number TI - Linearly constrained robust portfolio construction using Bauer-Householder bounds on the condition number UR - http://hdl.handle.net/11427/43840 ER - en_ZA
dc.identifier.urihttp://hdl.handle.net/11427/43840
dc.identifier.vancouvercitation, . Linearly constrained robust portfolio construction using Bauer-Householder bounds on the condition number. []. University of Cape Town ,Faculty of Science ,Department of Statistical Sciences, 2026 [cited yyyy month dd]. Available from: http://hdl.handle.net/11427/43840en_ZA
dc.language.rfc3066eng
dc.publisher.departmentDepartment of Statistical Sciences
dc.publisher.facultyFaculty of Science
dc.publisher.institutionUniversity of Cape Town
dc.subjectMean-variance optimization
dc.subjectBauer and Householder inequality
dc.subjectKantorovich inequality
dc.subjectgearing constraints
dc.subjectrobust optimization
dc.titleLinearly constrained robust portfolio construction using Bauer-Householder bounds on the condition number
dc.typeThesis / Dissertation
dc.type.qualificationlevelDoctoral
dc.type.qualificationlevelPhD
Files
Original bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
thesis_sci_2026_dalmeyer lara.pdf
Size:
4.93 MB
Format:
Adobe Portable Document Format
Description:
License bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
license.txt
Size:
1.72 KB
Format:
Item-specific license agreed upon to submission
Description:
Collections