FOREIGN EXCHANGE MARKET LINKAGES, 2017-2019

Brian D. Deaton
University of Arkansas at Monticello

ABSTRACT

Three causal search algorithms are used to find the causal structure underlying the six
most widely traded currencies (Australian dollar, Canadian dollar, euro, Great Britain pound
sterling, Japanese yen, and United States dollar) in the years 2017-2019. The linkages amongst
currencies found by each algorithm and within each year are illustrated as directed acyclic
graphs. Results show that the structure of the foreign exchange market is relatively stable over
time, but it is likely that there are latent variables that influence this market.
Keywords: foreign exchange, interdependence, causality, vector autoregression, causal search algorithm