Antoni Wiliński
Antoni Wiliński is a Polish professor (DSc, PhD) working at the intersection of computer science, machine learning and quantitative finance, with a long publication record on data-driven investment strategies. Over his career he has been affiliated with Koszalin University of Technology (Faculty of Electronics and Computer Science), the West Pomeranian University of Technology in Szczecin (Faculty of Computer Science and Information Technology), and — more recently — WSB Merito University in Gdańsk, in the Department of Finance and Management. His stated research interest is the extraction of knowledge from data using artificial-intelligence methods, with a particular focus on forecasting tasks in financial markets and the modelling of market-participant behaviour.
In this vault he appears as the author of Wilinski 2019 — “Time series modeling and forecasting based on a Markov chain with changing transition matrices” (Expert Systems with Applications, Vol. 133, 2019) — the only peer-reviewed source in the discrete-Markov price-prediction cluster that reports actual trading profit. The paper proposes a heterogeneous Markov Chain Trading Model whose transition matrix is re-estimated over a sequence of sliding windows, explicitly to combat the Non-Stationary Transition Matrix problem, with window length, window count and interval count tuned by machine learning. It belongs to a broader body of his work on machine-learning-adapted algorithmic-trading rules — including a 2014 study of investment strategies based on a classifier with rules adapted by machine learning, the Wiliński & Kovalerchuk (2017) Cognitive Systems Research paper on visual knowledge discovery for investment strategy, and later moving-average pattern-extraction work with Sochanowski and Nowicki (2022). He appears here as the named researcher behind the most explicit profitability claim made for the Markov-chain trading model — a claim this vault grades as weak because it rests on machine-learning-tuned parameters with no disclosed out-of-sample, net-of-cost track record or benchmark.
Antoni Wiliński [proposes_model] Markov Chain Trading Model Antoni Wiliński [opposes] Non-Stationary Transition Matrix
Connections
- Wilinski 2019 — proposes_model, 2019, source: https://www.sciencedirect.com/science/article/abs/pii/S0957417419303033
- Markov Chain Trading Model — proposes_model, 2019, source: https://www.sciencedirect.com/science/article/abs/pii/S0957417419303033
- Non-Stationary Transition Matrix — contradicts, 2019, source: https://www.sciencedirect.com/science/article/abs/pii/S0957417419303033