David Imhof, Emanuel W Viklund, Martin Huber
arXiv 16 Jul 2025 · Econometrics
arXiv:2507.12369 · PDF · DOI · OpenAlex · Extracted main text
We propose a novel application of graph attention networks (GATs), a type of graph neural network enhanced with attention mechanisms, to develop a deep learning algorithm for detecting collusive behavior, leveraging predictive features suggested in prior research. We test our approach on a large dataset covering 13 markets across seven countries. Our results show that predictive models based on GATs, trained on a subset of the markets, can be effectively transferred to other markets, achieving accuracy rates between 80% and 90%, depending on the hyperparameter settings. The best-performing configuration, applied to eight markets from Switzerland and the Japanese region of Okinawa, yields an average accuracy of 91% for cross-market prediction. When extended to 12 markets, the method maintains a strong performance with an average accuracy of 84%, surpassing traditional ensemble approaches in machine learning. These results suggest that GAT-based detection methods offer a promising tool for competition authorities to screen markets for potential cartel activity.
appendix boundary found by appendix_titled_section at “Appendix A: Statistical features” · 84% of the source is main text. Read the extracted text to check this.
The works this paper leans on most, across its whole bibliography — not restricted to papers in our corpus. Ranked by composite intensity, which combines how often a work is mentioned, how many sections mention it, and how much of that falls in the main text rather than the appendix.
| Reference | Intensity | Mentions | Sections | Main text | |
|---|---|---|---|---|---|
| 1 | Aaltio, A., R. Buri, A. Jokelainen, and J. Lundberg, Complementary b… (2025) | 1.000 | 7 | 3 | 100% |
| 2 | Aryal, G., and M. F. Gabrielli, Testing for collusion in asymmetric… (2013) | 1.000 | 6 | 4 | 100% |
| 3 | Garcia, M. J. R., V. R. Montequin, F. O. Fernandez, and J. M. V. Bal… (2022) | 1.000 | 5 | 3 | 100% |
| 4 | Wallimann, H., M. Huber, and D. Imhof, A machine learning approach f… (2022) | 0.961 | 9 | 5 | 89% |
| 5 | Huber, M., D. Imhof, and R. Ishii, Transnational machine learning wi… (2022) self | 0.928 | 15 | 6 | 80% |
| 6 | Huber, M., and D. Imhof, Machine learning with screens for detecting… (2019) self | 0.888 | 10 | 4 | 70% |
| 7 | Imhof, D., and H. Wallimann, Detecting bid-rigging coalitions in dif… (2021) self | 0.874 | 5 | 2 | 100% |
| 8 | Imhof, D., Detecting bid-rigging cartels with descriptive statistics… (2020) self | 0.843 | 4 | 3 | 75% |
| 9 | Anysz, H., A. Foremny, J. Kulejewski, and A. Nical, Collusion and bi… (2018) | 0.811 | 4 | 2 | 100% |
| 10 | Huber, M., and D. Imhof, Flagging cartel participants with deep lear… (2023) self | 0.811 | 4 | 2 | 100% |
Showing the top 10 of 48 scored citations.