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.
Patrick Minford, 04 January 2015
David F. Hendry, Grayham E. Mizon, 18 June 2014
Many central banks rely on dynamic stochastic general equilibrium models – known as DSGEs to cognoscenti. This column – which is more technical than most Vox columns – argues that the models’ mathematical basis fails when crises shift the underlying distributions of shocks. Specifically, the linchpin ‘law of iterated expectations’ fails, so economic analyses involving conditional expectations and inter-temporal derivations also fail. Like a fire station that automatically burns down whenever a big fire starts, DSGEs become unreliable when they are most needed.
Charles A.E. Goodhart, Dimitri Tsomocos, 26 November 2009
Standard DSGE models do not include the possibility of default. This column says that makes them useless for analysing financial crises. It proposes explicitly incorporating default and money into the microfoundations of DSGE models so as to offer a new framework for monetary and regulatory policy analysis.
Paul De Grauwe, 19 November 2009
The extraordinary assumptions of macroeconomic models have left the outside world perplexed about what economists have been doing during the last few decades. This column contrasts the incongruous rational expectations top-down model with a bottom-up model where no individual is capable of understanding the full complexity of a market system. The bottom-up model creates correlations in beliefs that generate waves of optimism and pessimism. The latter produce endogenous business cycles akin to the Keynesian “animal spirits”.
Alan Kirman, 14 November 2009
How will economic theory emerge from the global crisis? This column says that representative agent models and the efficient markets hypothesis are assumptions that have persisted too long in the face of empirical evidence. It argues that economic theory is due for an overhaul but fears that economists will resist such change.