Data Trends for Investment Professionals


Foreign Exchange and Currency Data on Quandl

Quandl has a number of high-quality databases covering foreign exchange rates and currencies: spot and reference rates, trade-weighted rates; real and PPP-adjusted rates, and much more.  This post lists a few of the most popular of these databases.

There are a number of good sources for high quality historical foreign exchange and currency data on Quandl.


Central Bank Reference Rates

A few central banks publish reference exchange rates across markets. The best of these is probably the Bank of England. BOE publishes rates to convert over 40 global currencies to the US Dollar, Pound Sterling, and the Euro. The BOE also publishes historical exchange rates for currencies that have been absorbed into the Euro zone. BOE FX rates include daily, monthly, quarterly and annual averages.

The Federal Reserve also publishes a number of foreign exchange rates as part of its weekly H.10 data release. These rates are certified by FRB-NY and published by FRB-STL. FRED’s forex rates include daily, monthly and annual averages for over 30 currencies against the US Dollar.

The European Central Bank publishes reference exchange rates against the Euro for over 30 currencies. In addition to these reference rates, the ECB publishes trade-weighted effective exchange rates for each currency against the EER-40. Trade-weighted exchange rates are weighted average nominal exchange rates, with weights proportional to bilateral trade volumes. This calculation method aims to capture changes in export/import competitiveness caused solely by exchange rate variations. ECB trade-weighted exchange rates come in both nominal and real (CPI-deflated) flavours.

The Central Bank of Brazil publishes monthly average exchange rates against the US dollar for about 30 major currencies. Many other central banks publish exchange rates for their own national currency against the global majors (USD, EUR, JPY, GBP).


International Organization Comparison Rates

Large international organizations like the World Bank and the United Nations all publish currency exchange rates. These have the advantage that they often include not just the nominal exchange rate, but also variations like real exchange rates, PPP-adjusted exchange rates, trade-weighted exchange rates and so on. But they have the disadvantage that they tend to be updated rather infrequently: monthly at best, annual at worst.

Another interesting dataset in this category comes from the Economist magazine. The Economist publishes both nominal and PPP-adjusted exchange rates, for 60 countries against the dollar, euro, sterling, yen and yuan; the purchasing-power-parity adjustment is, famously, based on the price of a Big Mac hamburger in each country.


Private Sector Market Rates

Central bank and international organization reference rates have a few known flaws.  Timing is the most obvious one.  Exchange markets are sampled infrequently (daily at best; weekly or monthly more often).  This is sub-optimal for most professional financial applications.  Even when observations are taken daily, they are often indexed to an arbitrary timestamp (typically, noon); this makes it hard to integrate reference exchange rates with other financial market data such as stock prices, futures settlements and interest rates, all of which have well-defined closing hours. Finally, even when the data is accurate and synchronized, it is often published with a lag, making it useless for time-sensitive use cases.

Another flaw is in the granularity of the data. Reference rates do not include transaction metrics such as open, high, low, bid or ask; nor do they capture volume or position size information. Published precision also tends to fall short of what’s actually observable in the market.  Cross-rates are another source of imperfection. Most reference rate calculators obtain cross rates by simply dividing one USD rate by another.  This is not a reflection of true market conditions at all.

Private sector forex data providers, especially exchange publishers, currency brokers and conversion bureaus, aim to explicitly avoid these flaws.  They typically provide the most accurate, trustworthy and detailed foreign exchange data, sampled directly from market transactions, and carefully synchronized.  We are currently working towards offering a few such private-sector sources on Quandl.


Update: Premium Forex Data

We now have a “premium” forex database on Quandl:  Global Foreign Exchange Rates, provided by vendor Exchange Data International. This database is truly professional-grade: forex rates for 168 currencies, aggregated from major FX dealers and investment banks at precisely 18:00 GMT and available on Quandl by 20:00 GMT the same day.  If you need currency data for professional purposes, we recommend this database.

(Founded in 1994 and with offices in London, New York, Montreal, Mumbai and Morocco, EDI is one of the world’s leading providers of professional grade financial data. EDI’s customers include major investment banks, brokerages, hedge funds and top tier universities.)

(As with all data on Quandl, foreign exchange data whether free or premium can be accessed using all the Quandl conduits: our website, API, or our 25+ integrations for R, Python, Excel, Maple, Matlab, C, Ruby and more).


Fix This
Created with Sketch.