Reinforcement learning for Exchange Rate Forecasting
Ability to make better forecast of future exchange rates is highly valuable for investors exposed to assets or liabilities in foreign currencies. Most large organizations in financial and non-financial domain have to manage their foreign currency exposures for achieving desired financial returns and or minimize the risks. In this project we compared three different approaches to forecast medium to long term exchange rate forecasting. We evaluated performance of econometric models, time series models, and reinforcement learning agents using Q-learning algorithm. We tested the models on different currently baskets including EUR/USD, GBP/USD, SEK/EUR, and NOK/EUR using market data from 2000-2018 covering different economic conditions.
Rakesh Rana works as Lead Data Scientist at Nordea Life & Pensions, Sweden. His work focuses on applying AI to solve business problems and create customer value. Rakesh received his M.S. degree in Finance and PhD in Computer Science from Chalmers/University of Gothenburg, Sweden. His work and research interests revolve around using data science and machine learning algorithms mainly within the financial domain.