Dinggao Liu (University of British Columbia) to Present a Study Co-conducted with Prof. Robert Ślepaczuk

We are pleased to share the results of studies conducted by researchers from our Faculty in collaboration with scholars from around the world. This time, we invite you to attend the lecture „A Time-Varying Interpretable Multi-Scale Transformer for Exchange Rate Returns Forecasting and Investment Decision Support”, which will be delivered by Dinggao Liu (University of British Columbia).

The co-authors of the study are Prof. Robert Ślepaczuk (Department of Quantitative Finance and Machine Learning) and Zhenpeng Tang (Fujian Agriculture and Forestry University).

The seminar will take place on 1 December at 18:30 in room B002 (Faculty of Economic Sciences, Długa 44/50). A remote participation is also available (Zoom link: https://uw-edu-pl.zoom.us/j/91585591246?pwd=8FrBk2Il30OFS7CjOBuCTRoBcbPIU1.1. To receive the registration details please contact us at: X+JgdwG?$k6!-{1#h42'Ex]#[Iy6UT[.wAc!e6q&4Ns'12e).

The presentation will last approximately about 45 minutes, followed by a discussion.

All attendees are kindly asked to arrive or log in 5 minutes before the start. The abstract of the presentation is provided below.

---

Accurately forecasting daily exchange rate returns represents a longstanding challenge in international finance, as the exchange rate returns are driven by a multitude of correlated market factors and exhibit high-frequency fluctuations. This paper proposes EXFormer, a novel Transformer-based architecture specifically designed for forecasting the daily exchange rate returns. We introduce a multi-scale trend-aware self-attention mechanism that employs parallel convolutional branches with differing receptive fields to align observations on the basis of local slopes, preserving long-range dependencies while remaining sensitive to regime shifts. A dynamic variable selector assigns time-varying importance weights to 28 exogenous covariates related to exchange rate returns, providing pre-hoc interpretability. An embedded squeeze-and-excitation block recalibrates channel responses to emphasize informative features and depress noise in the forecasting. Using daily EUR/USD, USD/JPY, and GBP/USD returns, we conduct out-of-sample evaluations across five different sliding windows. The model consistently outperforms the random walk benchmark, improving directional accuracy by a statistically significant margin of 8.5-22.8%. Nearly one year of trading backtests shows that these statistical gains translate into economically meaningful performance, obtaining cumulative net returns of 7%, 19%, and 9% with Sharpe ratios exceeding 1.8 after transaction costs. The robustness checks further confirm the model’s superiority under high-volatility and bear-market regimes. EXFormer offers a practical, interpretable, and knowledge-driven tool into exchange rate dynamics for international investors, multinational firms, and policy institutions.