Low-rank completion and recovery of correlation matrices

dc.contributor.advisorOuwehand, Peter
dc.contributor.advisorMc Walter, Thomas
dc.contributor.authorRamlall, Chetan K
dc.date.accessioned2020-02-13T09:53:43Z
dc.date.available2020-02-13T09:53:43Z
dc.date.issued2019
dc.date.updated2020-02-13T09:53:18Z
dc.description.abstractIn the pursuit of efficient methods of dimension reduction for multi-factor correlation systems and for sparsely populated and partially observed matrices, the problem of matrix completion within a low-rank framework is of particular significance. This dissertation presents the methods of spectral completion and convex relaxation, which have been successfully applied to the particular problem of lowrank completion and recovery of valid correlation matrices. Numerical testing was performed on the classical exponential and noisy Toeplitz parametrisations and, in addition, to real datasets comprising of FX rates and stock price data. In almost all instances, the method of convex relaxation performed better than spectral methods and achieved the closest and best-fitted low-rank approximations to the true, optimal low-rank matrices (for some rank-n). Furthermore, a dependence was found to exist on which correlation pairs were used as inputs, with the accuracy of the approximations being, in general, directly proportional to the number of input correlations provided to the algorithms.
dc.identifier.apacitationRamlall, C. K. (2019). <i>Low-rank completion and recovery of correlation matrices</i>. (). ,Faculty of Commerce ,Division of Actuarial Science. Retrieved from http://hdl.handle.net/11427/31080en_ZA
dc.identifier.chicagocitationRamlall, Chetan K. <i>"Low-rank completion and recovery of correlation matrices."</i> ., ,Faculty of Commerce ,Division of Actuarial Science, 2019. http://hdl.handle.net/11427/31080en_ZA
dc.identifier.citationRamlall, C. 2019. Low-rank completion and recovery of correlation matrices.en_ZA
dc.identifier.ris TY - Thesis / Dissertation AU - Ramlall, Chetan K AB - In the pursuit of efficient methods of dimension reduction for multi-factor correlation systems and for sparsely populated and partially observed matrices, the problem of matrix completion within a low-rank framework is of particular significance. This dissertation presents the methods of spectral completion and convex relaxation, which have been successfully applied to the particular problem of lowrank completion and recovery of valid correlation matrices. Numerical testing was performed on the classical exponential and noisy Toeplitz parametrisations and, in addition, to real datasets comprising of FX rates and stock price data. In almost all instances, the method of convex relaxation performed better than spectral methods and achieved the closest and best-fitted low-rank approximations to the true, optimal low-rank matrices (for some rank-n). Furthermore, a dependence was found to exist on which correlation pairs were used as inputs, with the accuracy of the approximations being, in general, directly proportional to the number of input correlations provided to the algorithms. DA - 2019 DB - OpenUCT DP - University of Cape Town KW - actuarial science LK - https://open.uct.ac.za PY - 2019 T1 - Low-rank completion and recovery of correlation matrices TI - Low-rank completion and recovery of correlation matrices UR - http://hdl.handle.net/11427/31080 ER - en_ZA
dc.identifier.urihttp://hdl.handle.net/11427/31080
dc.identifier.vancouvercitationRamlall CK. Low-rank completion and recovery of correlation matrices. []. ,Faculty of Commerce ,Division of Actuarial Science, 2019 [cited yyyy month dd]. Available from: http://hdl.handle.net/11427/31080en_ZA
dc.language.rfc3066eng
dc.publisher.departmentDivision of Actuarial Science
dc.publisher.facultyFaculty of Commerce
dc.subjectactuarial science
dc.titleLow-rank completion and recovery of correlation matrices
dc.typeMaster Thesis
dc.type.qualificationlevelMasters
dc.type.qualificationnameMPhil
Files
Original bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
thesis_com_2019_ramlall_chetan_k.pdf
Size:
2.63 MB
Format:
Adobe Portable Document Format
Description:
License bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
license.txt
Size:
0 B
Format:
Item-specific license agreed upon to submission
Description:
Collections