Please note our general information on theses including the guidelines for writing a thesis and our hints for literature searches. In addition, please note that most theses supervised by the institute have an empirical part. We therefore recommend that students interested in writing a Master thesis attend the modeling and research courses offered by the finance institutes (Financial Modeling, Research in Finance).

 

 

Investing

 

Using large language models for trading

Can large language models such as ChatGPT be useful for trading stocks?  In this thesis, you shall test this with the help of a live application in which you repeatedly ask an LLM (or several LLMs) in an automated way for recommendations . As a comparison, you would also determine the success of standard trading strategies. Contact: Prof. Löffler

Literature to get started: Pelster, M. and Val, J., 2024. Can ChatGPT assist in picking stocks?. Finance Research Letters, 59, p.104786.

 

Momentum and its variants

The momentum strategy is one of the most widely known and studied investment strategies. Several modifications of the strategy have been suggested. In this thesis, you shall implement the momentum strategy as well as important modifications of the strategy for a major stock market. As part of this, you can also examine whether large language models such as ChatGPT are good at aggregating the different dimensions of momentum information. Contact: Prof. Löffler

Literature to get started: Büsing, Pascal, Hannes Mohrschladt, and Susanne Siedhoff. "Decomposing momentum: The forgotten component." Journal of Banking & Finance 168 (2024): 107292.

 

Predicting stock market returns - a new approach 

You find the description of this topic below, under “Asset Pricing”. Since stock market predictions can also be used for market timing strategies, this is also an investing topic.

 

SUSTAINABLE FINANCE

 

Informativeness of green bond Impact reports

Green bonds are bonds whose proceeds are used for projects that benefit the environment. Companies that issue green bonds are supposed to report about the use of proceeds and the environmental benefits achieved with the proceeds. The precise questions you shall work on will be discussed in a meeting. The literature mentioned below is not really about impact reports but it may help you become familiar with the field and it also references a small number of impact reports that you can look at in order to get an idea of how they look like. (Some of the links may no longer work, though, as companies reorganize their websites.)  As part of the analysis, you could also examine whether large language models such as ChatGPT are good at assessing the informativeness of impact reports. Contact: Prof. Löffler

Literature to get started: ElBannan, M.A. and Löffler, G., 2024. How effectively do green bonds help the environment?. Journal of Banking & Finance, 158, p.107051.

 

ASSET PRICING 

 

Predicting stock market returns - a new approach 

Predicting stock market returns is anything but easy. Time and again, forecasting models have been proposed, only to fail to deliver. A recent research paper outlines a new yet simple methodological approach. In this thesis, you are to assess whether the proposed method is robust over time and across different markets. Contact: Prof. Löffler

Literature to get started:  Li, K., Li, Y., Lyu, C. and Yu, J., 2025. How to dominate the historical average. The Review of Financial Studies, 38(10), pp.3086-3116.

 

 

RISK

Performance of banks's internal value-at-risk models

In this thesis you shall examine the performance of the internal value-at-risk models of a few selected banks. The data will be retrieved from the banks' annual reports as described in the paper referenced below.  Contact: Prof. Löffler

Literature to get started: Pérignon, C., Deng, Z.Y. and Wang, Z.J., 2008. Do banks overstate their Value-at-Risk?. Journal of Banking & Finance, 32(5), pp.783-794.