If an economist selects the wrong model to study a question, the results are also likely to be wrong. In this video, Bruce Hansen talks to Soumaya Keynes about how model selection and combination can be used for forecasting with small error. Model combination methods are suited for forecasting and policy evaluation. This video was recorded in March 2016 during the Royal Economic Society’s Annual Conference held at the University of Sussex.
Bruce Hansen, 03 June 2016
Joan Paredes, Javier Pérez, Gabriel Pérez-Quirós, 12 July 2015
Uncertainty about fiscal policies can be damaging for economic performance, as it affects decisions about consumption, investment, and savings. This column argues that it is possible to reduce such uncertainty. Even if governments’ fiscal plans turn out to be (purposely) wrong ex post, they can convey useful information. It is just a matter of using the appropriate learning device whereby government promises are confronted every quarter with reality (i.e. what the government is actually doing).
Michele Ca'Zorzi, Jakub Mućk, Michał Rubaszek, 13 February 2015
Notwithstanding the progress made in the field of exchange rate economics, we still know very little of what drives major currencies. This column argues that the best that one can do is to assume that currencies move to gradually restore (relative) purchasing power parity. Contrary to widely held beliefs, this is in general a much better strategy than to just assume that the exchange rate behaves like a random walk.
Jon Danielsson, 18 January 2015
The Swiss central bank last week abandoned its euro exchange rate ceiling. This column argues that the fallout from the decision demonstrates the inherent weaknesses of the regulator-approved standard risk models used in financial institutions. These models under-forecast risk before the announcement and over-forecast risk after the announcement, getting it wrong in all states of the world.
Patrick Minford, 04 January 2015
Out-of-sample forecasting tests are increasingly used to establish the quality of macroeconomic models. This column discusses recent research that assesses what these tests can establish with confidence about macroeconomic models’ specification and forecasting ability. Using a Monte Carlo experiment on a widely used macroeconomic model, the authors find that out-of-sample forecasting tests have weak power against misspecification and forecasting performance. However, an in-sample indirect inference test can be used to establish reliably both the model’s specification quality and its forecasting capacity.
Philippe Andrade, Richard Crump, Stefano Eusepi, Emanuel Moench, 23 December 2014
Expectations are critical for macroeconomics and financial markets. But the expectation-formation process is not well understood. This column discusses some empirical characteristics of forecast disagreement from professional forecasters in the US, and discusses the ‘information frictions’ that underlie the heterogeneity of expectations.
Christiane Baumeister, Lutz Kilian, 19 November 2014
Futures prices are a potentially valuable source of information about market expectations of asset prices. This column discusses a general approach to recovering this expectation when there is no agreement on the nature of the time-varying risk premium contained in futures prices. The authors illustrate this approach by tackling the long-standing problem of how to recover the market expectation of the price of crude oil.
Charles Goodhart, Philipp Erfurth, 04 November 2014
Most of the world is now at the point where the support ratio is becoming adverse, and the growth of the global workforce is slowing. This column argues that these changes will have profound and negative effects on economic growth. This implies that negative real interest rates are not the new normal, but rather an extreme artefact of a series of trends, several of which are coming to an end. By 2025, real interest rates should have returned to their historical equilibrium value of around 2.5–3%.
Giang Ho, Paolo Mauro, 12 September 2014
Forecasters often predict continued rapid economic growth into the medium and long term for countries that have recently experienced strong growth. Is this optimism warranted by past international growth experience? This column explores this question by looking at economic growth forecasts at longer-term horizons.
Jon Danielsson, Kevin James, Marcela Valenzuela, Ilknur Zer, 08 June 2014
Risk forecasting is central to financial regulations, risk management, and macroprudential policy. This column raises concerns about the reliance on risk forecasting, since risk forecast models have high levels of model risk – especially when the models are needed the most, during crises. Policymakers should be wary of relying solely on such models. Formal model-risk analysis should be a part of the regulatory design process.
Hites Ahir, Prakash Loungani, 14 April 2014
Forecasters have a poor reputation for predicting recessions. This column quantifies their ability to do so, and explores several reasons why both official and private forecasters may fail to call a recession before it happens.
Barbara Rossi, 14 November 2013
Predicting exchange rates is still an inexact science. Economic models perform poorly, and a plethora of alternative methods have been attempted. This column guides the reader through the state of the art, reviewing various predictors, models, and data specifications. Despite a large and divergent literature chasing this holy grail, the toughest benchmark remains the random walk without drift.
Rossana Merola, Javier Pérez, 01 May 2013
Who should we trust when it comes to fiscal forecasts: governments or independent agencies? This column argues that this question is, in fact, a red herring: empirical evidence suggests that in the past, international agencies’ fiscal forecasts were partially affected by the same problems that the literature widely acknowledges for governmental forecasts. An attractive solution is independent national forecasters.
Peter Tillmann, 23 February 2012
As the US Federal Reserve starts to increase the transparency of its decision-making process, including the release of economic forecasts and interest-rate projections, this column asks whether these projections reflect strategic motives that might make them less accurate and less useful to those wanting to predict monetary policy.
Volker Wieland, Maik Wolters, 13 February 2012
Where were economists when the global recession hit? Or rather, where were their forecasts in the years before? This column argues that clearly some of the models were at fault. To correct this, it proposes a ‘comparative approach’ to macroeconomic analysis where models compete for the right to be taken seriously.
Lutz Kilian, 23 June 2011
Reduced Libyan output, broader political unrest in the Middle East, and a slow global recovery have raised the uncertainty surrounding oil prices. This column discusses the challenges and value of forecasting future oil prices in real time, as opposed to fitting models to revised oil prices released months after economic decisions are made.
Refet Gürkaynak, Rochelle Edge, 28 February 2011
Studies have shown that the forecasts from dynamic stochastic general equilibrium models perform better than central banks' judgemental forecasts as well as forecasts based on statistical analysis but without a theoretical foundation. This column shows that performing better is hardly good performance given how badly all three forecasts compare with reality.
James Reade, David Hendry, 11 June 2009
A vital challenge confronting economists is how to forecast, especially during a recession because livelihoods depend on those forecasts. This column discusses choosing amongst forecasts and outlines the concerns involved in averaging across models or using general-to-specific model searches.
Hal Varian, 08 May 2009
Hal Varian, on leave from the University of California, Berkeley, talks to Romesh Vaitilingam about the kinds of things he thinks about as chief economist at Google – using the tools of economics to analyse business opportunities, making money out of search and content, and forecasting (including Google’s flu forecasts). The interview was recorded at the American Economic Association meetings in San Francisco in January 2009.
Riccardo Cristadoro, Alessandro Secchi, Giovanni Veronese, 22 November 2008
We are in the most serious financial crisis since WWII, and many did not see it coming until it was too late. This column contrasts the performance of €-coin, a real-time monthly indicator for the euro area, with that of professional forecasters. €-coin seems to provide a more accurate assessment of current economic trends. Unfortunately, it is at its lowest level since 1993.