Proposals for Theses

Please note our general information on theses. 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).

ASSET PRICING

 

Machine learning and stock market predictability

You shall examine whether estimation approaches from machine learning can help improve the precision of stock market forecasts compared to existing approaches. The paper stated below is an example of an "existing approach". The choice of machine learning methods will be discussed before you start working on the thesis. (Contact: Prof. Löffler).

Literature to get started: Rapach, D. E., Strauss, J. K., & Zhou, G. (2010). Out-of-sample equity premium prediction: Combination forecasts and links to the real economy. Review of Financial Studies23(2), 821-862.

 

INVESTING

 

Prospect theory and predictability

A recent paper has shown that prospect theory can help predict the cross-section of stock returns. In this thesis - after a review of the literature on prospect theory and asset pricing - you shall employ the approach of this paper to examine whether prospect theory can help predict returns of international stock market indices (Contact: Prof. Löffler).

Literature to get started:  Barberis, Mukherjee, Wang (2016): Prospect Theory and Stock Returns: An Empirical Test. Review of Financial Studies 29, 3068-3107.

  

RISK

Sentiment and Credit Risk

The asset pricing literature has found that sentiment, mood or whatever you call it can impact prices and investor behavior. Decisions related to credit risk may also be affected by sentiment but there is not much literature on it. In this thesis, you shall first review the literature and then conduct an empirical study on the links between sentiment and credit risk (Contact: Prof. Löffler)

Literature to get started:  Agarwal, S., Duchin, R., & Sosyura, D. (2012). In the mood for a loan: The causal effect of sentiment on credit origination. Working Paper, SSRN.

 

Machine learning approaches for credit scoring

Credit scoring models are models that predict defaults. In this thesis, you shall review the literature on estimation approaches and then conduct your own empirical study in which you compare the standard logit approach with a small number (1-3) of machine learning approaches (Contact: Prof. Löffler)

Literature to get started: Lessmann, S., Baesens, B., Seow, H. V., & Thomas, L. C. (2015). Benchmarking state-of-the-art classification algorithms for credit scoring: An update of research. European Journal of Operational Research, 247(1), 124-136.

 

Predicting default rates with machine learning

The risk of a given credit rating such as BBB varies with the business cycle. It is therefore useful to have a model that predicts default rates for rating categories. In this thesis, you shall review the literature on default rate prediction and then conduct your own empirical study in which you use approaches from statistical learning / machine learning in addition to standard ones (Contact: Prof. Löffler)

Literature to get started: Chapter "Prediction of Default and Transition Rates" in Löffler, G., & Posch, P. N. (2011). Credit risk modeling using Excel and VBA. John Wiley & Sons.