Currency Data
About the Paper
The dataset, described below, accompanies the following paper:
Della Corte, Pasquale and Steven J. Riddiough (2025), “Currency Speculation”, Working Paper, Imperial College London and the University of Toronto.
​
This paper has been prepared for publication in the Annual Review of Financial Economics. The working paper version is available here. The BibTeX file can be downloaded here.
​
The data are freely available for academic use. If you use them, we kindly ask that you cite the paper.​​
​
We intend to update this dataset at regular intervals to reflect newly released data.​
About the Dataset
Each file reports excess returns for the currency samples fully described in the paper:
-
All Currencies (All)
-
Open Capital Account Currencies (Open)
-
Floating Regime Currencies (Floating)
-
High Turnover Currencies (High)
-
Developed and Emerging (G20)
-
Developed (G10)
​​
We provide two versions of excess returns:
-
Gross Excess Returns - returns computed using mid-quoted spot and forward exchange rates
-
Net Excess Returns - returns adjusted for effective bid–ask spreads, set to 25% of the actual quoted spread.
We rank currencies at the end of each month t using a pre-determined sorting variable, form a given number of baskets, and then calculate the equally-weighted average excess return for each basket between months t and t + 1. The Dollar and Dollar Carry strategies use a single basket, while all other strategies are based on five portfolios. To make our strategies more realistic, the sorting variable is lagged by one business day. For instance, in March 2025, we measure the sorting signal on March 27, form the portfolios on March 28, and calculate their average excess return on April 30.
We report the full set of portfolios across strategies (carry, value, momentum, etc.) and the corresponding long/short factors, constructed as the return on the highest-ranked portfolio minus the return on the lowest-ranked portfolio.
Below, you can download the data using the symbol or preview them in Google Sheets using the symbol ​ .
