Modern portfolio optimisation under regime switching

Simple item page
dc.contributor.advisorAlfeus, Mesiasen_ZA
dc.contributor.authorSteenkamp, Cara Yvetteen_ZA
dc.contributor.otherStellenbosch University. Faculty of Economic and Management Sciences. Dept. of Statistics and Actuarial Science.en_ZA
dc.date.accessioned2022-03-04T12:54:39Z
dc.date.accessioned2022-04-29T09:40:15Z
dc.date.available2022-03-04T12:54:39Z
dc.date.available2022-04-29T09:40:15Z
dc.date.issued2022-04
dc.descriptionThesis (MCom)--Stellenbosch University, 2022.en_ZA
dc.description.abstractENGLISH SUMMARY: The main objective of this assignment is to consider modern portfolio optimisation under regimes. Unobservable regimes are assumed to be modulated by a time-change Markov process. These models are well-known as Markov regime switching models. The Markov regime switching models are applied to portfolios that consist of a lagged model and a factor model. The lagged model represents a portfolio of 20 stocks which have been lagged by a day and then classified into regimes whereas the factor model uses 5 different global risk factors or measures namely, the 3 Fama-French factors, VIX and a spread between the 3-month JIBAR rate and SAFEX overnight rate to estimate the unobserved regimes for the portfolio. This assignment considers two regimes. Two regimes are classified representing the bull and bear markets, periods when the financial market is doing well and when the market is on a downturn respectively. Thereafter, optimisation is performed by taking the estimated regimes into account and obtaining the optimal portfolio allocations. Optimisation methods such as Sharpe ratio method and risk budget method are investigated. For each of these optimisation methods the portfolios were rebalanced to evaluate the financial markets at the start of the new investment period, classify it either into a new regime or remaining in the current state and then adjusting the portfolio weights. Portfolio optimisation including the regimes are then compared to classical modern portfolio optimisation without regimes consideration. Results show that portfolio optimisation with regimes obtained the highest Sharpe ratio, indicating the economic benefit of inclusion of regime switching characteristics in modern portfolio optimisation.en_ZA
dc.description.abstractAFRIKAANSE OPSOMMING: Die hoof doel van die opdrag is om die moderne portefeulje optimalisering onder regimes te oorweeg. Dit word aangeneem dat onwaarneembare regimes gemoduleer word deur die tydveranderende Markov proses. Hierdie modelle is welbekend as Markov regime oorskakelings modelle. Die Markov regime oorskakelings modelle word toegepas op ‘n portefeulje wat bestaan uit ‘n sloer model en ‘n faktor model. Die sloer model verteenwoordig ‘n portefeulje van 20 aandele wat vir ‘n dag sloer en dan eers geklassifiseer word in regimes waar die faktor model 5 verskillende globale risiko faktore gebruik of meet naamlik, die 3 Fama-French faktore, VIX en die verspreiding tussen die 3-maande JIBAR koers en SAFEX oornag koers om te bepaal wat die onwaarneembare regimes vir die potefeulje is. Hierdie opdrag oorweeg twee regimes. Twee regimes word geklassifiseer deur die bul en die beer markte, periodes wanneer die finansi¨ele mark opwaarts neig en wanneer die mark afwaards neig. Daarna word optimalisering gedoen deur die skatting van die regimes in ag te neem en die optimale portefeulje allokasies te verkry. Optimalisering metodes soos die Sharp verhouding metode en risiko begroting metode word nagevors. Vir elkeen van die optimalisering metodes is die portefeuljes weer gebalanseer om die finansi¨ele mark te evalueer aan die begin van die nuwe bellegings periode. Dit word dan geklassifiseer in ‘n nuwe regime, of dit bly dieselfde en die portefeulje gewig word daar volgens aangepas. Portefeulje optimalisering insluitend die regimes word dan vergelyk met die klassieke moderne portefeulje optimalisering sonder om die regimes in ag te neem. Resultate wys dat portefeulje optimalisering wat regimes insluit, verkry die hoogste Sharpe verhouding, dit dui die ekonomiese voordeel om die regime oorskakelings eienskappe in ‘n moderne portefeulje optimalisering.af_ZA
dc.description.versionMasters
dc.identifier.urihttp://hdl.handle.net/10019.1/124903
dc.language.isoen_ZAen_ZA
dc.publisherStellenbosch : Stellenbosch University
dc.rights.holderStellenbosch University
dc.subjectPortfolio management -- South Africaen_ZA
dc.subjectInvestment management -- South Africaen_ZA
dc.subjectInvestment analysis -- South Africaen_ZA
dc.subjectBusiness enterprises -- Financeen_ZA
dc.subjectUCTD
dc.titleModern portfolio optimisation under regime switchingen_ZA
dc.typeThesisen_ZA
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
steenkamp_modern_2022.pdf
Size:
1.24 MB
Format:
Adobe Portable Document Format
Description:
Download
Collections
Masters Degrees (Statistics and Actuarial Science)