Computer Science Thesis Topics in Computational Finance
Summary
The document discusses ways to frame computational finance as a computer science master's thesis. Suggested directions include building investment recommendations from time-series regression models, evaluating predictive performance with machine learning on historical high-frequency foreign-exchange data, and studying the software environment, trading simulations, and algorithms used in the research.
Other examples focus on the computational implementation of numerical methods, including radial basis function approximations for high-dimensional option pricing and differential-equation methods for option valuation. These projects can emphasize algorithm design, accuracy, and speed. The responses indicate that suitability depends on the institution's degree requirements; they offer examples rather than a general approval rule, and provide no comparative evaluation of the proposed topics or evidence of trading profitability.
Key ideas
- A computational finance thesis can center on algorithms and their implementation.
- Regression and machine learning on financial time series can support a trading-system project.
- Option-pricing research can focus on numerical accuracy and computational speed.
- Thesis suitability depends on the requirements of the degree-granting institution.
Tags
Full text
# Is it possible to defend a Computer Science master thesis by writing a project about quantitative finance? # Is it possible to defend a Computer Science master thesis by writing a project about quantitative finance? What are features and examples of computational finance (financial computing) problems (for thesis project in Master in Computer Science)? Is it possible to defend Computer Science master thesis by writing project about quantitative finance? As far as I have read articles in arxiv.org Computational Finance part, then most of them is about numerical methods and they are more relevant to mathematics than computer science. What can be good theme for the project? ## Answer by Svisstack (score 2) https://quant.stackexchange.com/a/27726 My CS master was: Investment recommendations generated using prediction models based on regression in time series Description: The purpose of this paper is to present the environment for generating investment recommendations from predictive models based on regression and evaluating their performance using machine learning techniques on historical high frequency data obtained from the foreign exchange market. The work also can find basic information about trading platforms, simulation environments and scientific information related to techniques, models and algorithms used in the work or forming an alternative or expansion for solutions applied. Probably can't post it on arxiv.org because it was not in English language. ## Answer by Chris Degnen (score 1) https://quant.stackexchange.com/a/27727 You could focus on the algorithms used to implement the numerical methods. ## Answer by millovanovic (score 1) https://quant.stackexchange.com/a/31358 This mostly depends on the regulations of the institution that is issuing a degree. I have been mentoring myself several CS master theses focused on Computational Finance, mainly related to solving high-dimensional option pricing problems using Radial Basis Function approximations. Therefore, it is possible. ## Answer by sen_saven (score 1) https://quant.stackexchange.com/a/31377 My CS master was: "Option Pricing through differential Equations". It had mostly focused on the numerical algorithms used to pricing and their accuracy/speed. I doubt you would have any issue defending your thesis if you pick something similar...
Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.