Bayesian forecasting of stock returns using simultaneous graphical dynamic linear models

Simple item page
dc.contributor.advisorBartlett, Bruceen_ZA
dc.contributor.advisorBecker, Ronnieen_ZA
dc.contributor.authorKyakutwika, Nelsonen_ZA
dc.contributor.otherStellenbosch University. Faculty of Science. Dept. of Applied Mathematics.en_ZA
dc.date.accessioned2022-11-22T09:58:19Z
dc.date.accessioned2023-01-16T12:51:31Z
dc.date.available2022-11-22T09:58:19Z
dc.date.available2023-01-16T12:51:31Z
dc.date.issued2022-12
dc.descriptionThesis (MSc) -- Stellenbosch University, 2022.en_ZA
dc.description.abstractENGLISH ABSTRACT: Cross-series dependencies are crucial in obtaining accurate forecasts when forecast- ing a multivariate time series. Simultaneous Graphical Dynamic Linear Models (SGDLMs) are Bayesian models that elegantly capture cross-series dependencies. This study aims to forecast returns of a 40-dimensional time series of stock data using SGDLMs. The SGDLM approach involves constructing a customised dy- namic linear model (DLM) for each univariate time series. Every day, the DLMs are recoupled using importance sampling and decoupled using mean-field varia- tional Bayes. We summarise the standard theory on DLMs to set the foundation for studying SGDLMs. We discuss the structure of SGDLMs in detail and give de- tailed explanations of the proofs of the formulae involved. Our analyses are run on a CPU-based computer; an illustration of the intensity of the computations is given. We give an insight into the efficacy of the recoupling/decoupling techniques. Our results suggest that SGDLMs forecast the stock data accurately and respond to market gyrations nicely.en_ZA
dc.description.abstractAFRIKAANSE OPSOMMING: Kruisreeksafhanklikhede is van kardinale belang om akkurate voorspellings te ver- kry wanneer ’n meervariant tydreeks voorspel word. Gelyktydige grafiese dina- miese lineeˆre modelle (SGDLMs) is Bayesiaanse modelle wat kruisreeksafhanklik- hede elegant vasleˆ. Hierdie studie het ten doel om opbrengste van ’n 40-dimensionele tydreeks van voorraaddata met behulp van SGDLMs te voorspel. Die SGDLM- benadering behels die konstruksie van ’n pasgemaakte dinamiese lineeˆre model (DLM) vir elke eenvariant tydreeks. Elke dag word die DLM’s herkoppel met be- hulp van belangrikheidsteekproefneming en ontkoppel met behulp van gemiddelde- veld variasie Bayes. Ons som die standaardteorie oor DLM’s op om die grondslag te leˆ vir die bestudering van SGDLM’e. Ons bespreek die struktuur van SGDLM’e in detail en gee gedetailleerde verduidelikings van die bewyse van die betrokke formules. Ons ontledings word op ’n SVE-gebaseerde rekenaar uitgevoer; ’n il- lustrasie van die intensiteit van die berekeninge word gegee. Ons gee ’n insig in die doeltreffendheid van die herkoppeling/ontkoppelingstegnieke. Ons resultate dui daarop dat SGDLM’s die voorraaddata akkuraat voorspel en mooi reageer op markwisselings.af_ZA
dc.description.versionMastersen_ZA
dc.format.extentxi 90 pages : illustrationsen_ZA
dc.identifier.urihttp://hdl.handle.net/10019.1/126128
dc.language.isoen_ZAen_ZA
dc.publisherStellenbosch : Stellenbosch Universityen_ZA
dc.rights.holderStellenbosch Universityen_ZA
dc.subjectSimultaneous graphical dynamic linear modelsen_ZA
dc.subjectBayesian -- Analysisen_ZA
dc.subjectStock returns -- Data processingen_ZA
dc.subjectDynamic linear modelsen_ZA
dc.subjectRecoupling modelsen_ZA
dc.subjectDecoupling modelsen_ZA
dc.subjectStock exchanges -- Data processingen_ZA
dc.subjectUCTDen_ZA
dc.titleBayesian forecasting of stock returns using simultaneous graphical dynamic linear modelsen_ZA
dc.typeThesisen_ZA
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
kyakutwika_bayesian_2022.pdf
Size:
1.53 MB
Format:
Adobe Portable Document Format
Description:
Download
Collections
Masters Degrees (Applied Mathematics)