Navigation Enabled with membership or trial


Executive Summary
 

Print Individual
Sections or
Entire Issue

 

Navigation Enabled with membership or trial


Currencies
(3, 6, 12-months)
 
Summary of currency forecasts (3, 6 & 12 months)
 
Detailed currency forecasts (3, 6 & 12 months)

Interest rate forecasts

Growth forecasts

Inflation forecasts

Current account forecasts
 
Forecasts for minor currencies

Long-term currency forecasts(1-5 years)


 

Navigation Enabled with membership or trial


Recent developments
& outlook for 32 countries

Latest weekly update


 
Forecasting Risk & Risk Tolerance
 
List of Contributors
 
Track record
 
Senior personnel
 
How financial officers use FX4casts effectively
 
Advantages of FX4casts.com™
 
FX rates
yr. end & averages
2000-2009

© 2010

 




 

Three studies confirm FX4casts superior track record

·       Survey Data on Exchange Rate Expectations.  Professors Jeffrey Frankel, Harvard, and Menzie Chinn, University of California. “We find some ability in the survey data to predict future movements in the exchange rate (and in the right direction!)”  The empirical results in this paper are based on data from FX4casts, previously The Financial Times Currency Forecasters’ Digest. In Monetary Policy, Capital Flows and Financial Market Developments in the Era of Financial Globalization, edited by William Allen and David Dickinson.

·       Predictability in Financial Markets: What do Survey Expectations Tell Us? National Bureau of Economic Research by Philippe Bacchetta and Elmar Mertens, Study Center Gerzensee, University of Lausanne and Eric van Wincoop, University of Virginia. This study uses FX4casts consensus currency forecasts from August 1986 through July 2005. It found there was “extensive evidence of predictability.”

·    Returns to Following Currency Forecasts, Patrick Lehner, Swiss Federal Institute of Technology, Zurich, John Okunev and Derek White, Principal Global Investors. This study evaluates FX4casts consensus currency forecasts from 1984-2003. The summary says: “By focusing on the extreme currencies, it was found that from 1995 through 2003, a return of 57 basis points a month could be earned…Returns earned by following the consensus recommendations increased greatly when technical and/or interest rate signals give similar rankings. For example, when twelve-month currency forecasts agree with technical and interest rate signals, the average monthly return jumped to 174 basis points.”
The complete study is on www.ssrn.com.

Subscriber Comments

“By providing a consensus of expectations, we have a higher level of comfort in FX4casts than forecasts of individual organizations, which vary widely.”   
                                    -
Joyce King-Lavinder, Treasury Department, Coca-Cola

“FX4casts has an excellent track record.”
                         -
Deborah L. Erickson, International Treasury Manager, Exxon Mobil

Similar studies – similar results

·       “Studies have repeatedly shown that combining predictions from diverse sources and methods generally results in significant gains in accuracy.”

Geoffrey Moore, National Bureau of Economic Research, Inc.

·    “Consensus forecasting has emerged as the most accepted method of evaluating the financial, economic and currency outlook. The consensus has proved to be more accurate than the individual forecasts comprising it.”

Walter Hoadley, Senior Research Fellow, Hoover Institute.

·    “At one time or another, every professional forecaster is wrong. Taken as a group, however, forecasters have a good record. Reliance on a consensus can at least keep you aware of signs that currency trends are changing.”

Peter L. Bernstein and Theodore, Harvard Business Review 

FX4casts improves accuracy with the geometric mean

FX4casts uses the geometric mean when computing consensus currency forecasts. By exponentially reducing the weight given to extreme forecasts, this reflects more accurately the contributors’ predominant expectations. The arithmetic mean, by contrast, is often misleading. For example, the average of 1, 2, 3 and 10 is 4. If the 10 is given equal weight, the arithmetic mean  would not reflect the group’s primary expectations as well as would the geometric mean.

 

RETURN TO TOP OF PAGE
FX4casts.com™