Full Text Available
Note: Clicking the button above will open the full text document at the original institutional repository in a new window.
A walk-forward multi-factor machine learning investment process.
Thesis (PhD)--Stellenbosch University, 2024.
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Published: |
Stellenbosch : Stellenbosch University
2025
|
| Subjects: | |
| Tags: |
No Tags, Be the first to tag this record!
|
| _version_ | 1867614099002621952 |
|---|---|
| access_status_str | Open Access |
| author | Wessels, Zander |
| author2 | Engelbrecht, Andries P. |
| author_browse | Engelbrecht, Andries P. Wessels, Zander |
| author_facet | Engelbrecht, Andries P. Wessels, Zander |
| author_sort | Wessels, Zander |
| collection | Thesis |
| dc_rights_str_mv | Stellenbosch University |
| description |
Thesis (PhD)--Stellenbosch University, 2024. |
| format | Thesis |
| id | oai:scholar.sun.ac.za:10019.1/131970 |
| institution | Stellenbosch University (South Africa) |
| last_indexed | 2026-06-10T12:46:39.009Z |
| license_str | Other — see source repository |
| provenance_str_mv | Harvested via OAI-PMH from SUNScholar — Stellenbosch University Repository |
| publishDate | 2025 |
| publishDateRange | 2025 |
| publishDateSort | 2025 |
| publisher | Stellenbosch : Stellenbosch University |
| publisherStr | Stellenbosch : Stellenbosch University |
| record_format | dspace |
| source_str | SUNScholar — Stellenbosch University Repository |
| spelling | oai:scholar.sun.ac.za:10019.1/131970 A walk-forward multi-factor machine learning investment process. Wessels, Zander Engelbrecht, Andries P. Stellenbosch University. Faculty of Engineering. Dept. of Industrial Engineering. Investments -- Mathematical models Machine learning Artificial intelligence -- Economic aspects UCTD Thesis (PhD)--Stellenbosch University, 2024. ENGLISH ABSTRACT: It can be said that traditional asset management modelling lacks true empiricism due to the difficulty of accurate simulation of investment strategies through time. This is often demonstrated by positively biased academic results that do not reflect real-world outcomes. This thesis addresses this problem by developing a walk-forward platform that is inspired by the blackboard-expert architecture and that can simulate investment processes through time, providing reliable repeatability, which is the essence of good science. The proposed platform accounts for common biases, such as survivorship bias and forwardlooking bias, to create a true hypothesis-testing engine. When considering the investment process as it pertains to equities, this thesis argues that the process can be fully closed under the questions of “what should I buy?”, “when should I buy?”, and “how much should I buy?”. The thesis then aims to use the hypothesis engine to build an example investment pipeline that can test and automate a whole investment process. The proposed investment pipeline will answer the questions of “what”, “when”, and “how much”, respectively. Traditional factor models and machine learning models for stock selection will be explored to answer the “what” question. The thesis argues that training models using the proposed engine is the correct way to do so in a non-stationary time series setting. The “how much” question will be addressed using portfolio optimisation, with particular consideration given to particle swarm optimisation. The “when” question will be briefly discussed as a further research idea. AFRIKAANSE OPSOMMING: Daar kan geargumenteer word dat ware empirisme ontbreek in tradisionele batebestuursmodellering as gevolg van die moeisaamheid om beleggingsstrategieë akkuraat deur tyd te simuleer. Dit word dikwels gedemonstreer deur positief bevooroordeelde akademiese resultate wat nie werklike uitkomste weerspieël nie. Hierdie tesis spreek hierdie probleem aan deur ’n platform te ontwikkel wat geïnspireer is deur die “blackboard-expert” argitektuur en wat beleggingsprosesse deur tyd kan simuleer. Sodoende word die doelwit van betroubare herhaalbaarheid, wat die wese van goeie wetenskap is, bereik. Die voorgestelde platform neem algemene vooroordele, soos oorlewingsvooroordeel en vooruitskouende vooroordeel, in ag om ’n ware hipotese-toetsing enjin te skep. Wanneer die beleggingsproses met betrekking tot aandele oorweeg word, argumenteer hierdie tesis dat die proses ten volle gesluit kan word onder die vrae “wat moet ek koop?”, “wanneer moet ek koop?” en “hoeveel moet ek koop?”. Die tesis beoog dan om die hipotese-enjin te gebruik om ’n voorbeeldbeleggingspyplyn te bou wat ’n hele beleggingsproses kan toets en outomatiseer. Die voorgestelde beleggingspyplyn sal onderskeidelik die vrae van “wat”, “wanneer” en “hoeveel” beantwoord. Tradisionele faktormodelle en masjienleermodelle vir aandele seleksie sal ondersoek word om die “wat” vraag te beantwoord. Die tesis argumenteer dat die passing van modelle met behulp van die voorgestelde enjin die korrekte manier is om dit te doen in ’n niestasionêre tydreeks konteks. Die “hoeveel” vraag sal aangespreek word deur portefeulje-optimering, met besondere oorweging gegee aan swerm-optimering. Die “wanneer” vraag sal kortliks bespreek word as ’n verdere navorsingsidee. Doctoral 2025-05-02T13:17:21Z 2025-05-02T13:17:21Z 2024-12 Thesis https://scholar.sun.ac.za/handle/10019.1/131970 Stellenbosch University x, 357 pages : illustrations application/pdf Stellenbosch : Stellenbosch University |
| spellingShingle | Investments -- Mathematical models Machine learning Artificial intelligence -- Economic aspects UCTD Wessels, Zander A walk-forward multi-factor machine learning investment process. |
| title | A walk-forward multi-factor machine learning investment process. |
| title_full | A walk-forward multi-factor machine learning investment process. |
| title_fullStr | A walk-forward multi-factor machine learning investment process. |
| title_full_unstemmed | A walk-forward multi-factor machine learning investment process. |
| title_short | A walk-forward multi-factor machine learning investment process. |
| title_sort | walk forward multi factor machine learning investment process |
| topic | Investments -- Mathematical models Machine learning Artificial intelligence -- Economic aspects UCTD |
| url | https://scholar.sun.ac.za/handle/10019.1/131970 |
| work_keys_str_mv | AT wesselszander awalkforwardmultifactormachinelearninginvestmentprocess AT wesselszander walkforwardmultifactormachinelearninginvestmentprocess |