Francesca Monti

Forecast with judgment and models

This paper proposes a simple and model-consistent method for combining forecasts generated by structural micro-founded models and judgmental forecasts. The method also enables the judgmental forecasts to be interpreted through the lens of the model. We illustrate the proposed methodology with a real-time forecasting exercise, using a simple neo-Keynesian dynamic stochastic general equilibrium model and prediction from the Survey of Professional Forecasters

Incorporating conjunctural analysis in structural models

This volume celebrates the work of Michael Woodford and his many contributions to economics. One of Mikes most influential papers is the 1997 paper (co-authored with Julio Rotemberg) An optimization-based econometric framework for the evaluation of monetary policy. This paper constituted the first attempt at estimation of a small scale dynamic stochastic general equilibrium model (DSGE) in which prices are set by monopolistically competitive firms, and prices cannot be instantaneously and costlessly adjusted. Since the work of Rotemberg and Woodford, these models have become more complex and increasingly large [see Christiano, Eichenbaum and Evans (2005), Smets and Wouters (2003), and, more recently, Christoffel, Coenen and Warne (2008) and Adolfson, Laseen, Linde and Svensson (2008)]. By explicitly taking into account forwardlooking behavior on the part of the agents, DSGEs provide a useful framework to analyze the effects of alternative policies. These models are now routinely used in many central banks, including the European Central Bank, and knowledge has been built up on their reliability, their forecasting performance and on what are the reasonable values for calibrated parameters and the setting of the priors.

Ambiguity, Monetary Policy and Trend Inflation

We develop a model that can explain the evolution of trend inflation in the United States in the three decades before the Great Recession as a function of the reduction in uncertainty about the monetary policy maker’s behaviour. The model features ambiguity-averse agents and ambiguity regarding the conduct of monetary policy, but is otherwise standard. Trend inflation arises endogenously and has these determinants: the strength with which the central bank responds to inflation, the degree of uncertainty about monetary policy perceived by the private sector, and, if it exists, the inflation target. Given the importance of monetary policy for the determination of trend inflation, we also study optimal monetary policy in the case of lingering ambiguity.

A Bayesian VAR benchmark for COMPASS

We estimate a Bayesian VAR analogue to the Bank of England’s DSGE model (COMPASS) and assess their relative performance in forecasting GDP growth and CPI inflation in real time between 2000 and 2012. We find that the BVAR outperformed COMPASS when forecasting both GDP and its expenditure components. In contrast, the performance of these models was similar when forecasting CPI. We also find that, despite underpredicting inflation at most forecast horizons, the BVAR density forecasts outperformed those of COMPASS. Both models overpredicted GDP growth at all forecast horizons, but the BVAR outperformed COMPASS at forecast horizons up to one year ahead. The BVAR’s point and density forecast performance is also comparable to that of a Bank of England in-house statistical suite for both GDP and CPI inflation and to the Inflation Report projections. Our results are broadly consistent with the findings of similar studies for other advanced economies.

Can a Data-Rich Environment Help Identify the Sources of Model Misspecification?

This paper proposes a method for detecting the sources of misspecification in a DSGE model based on testing, in a data-rich environment, the exogeneity of the variables of the DSGE with respect to some auxiliary variables. Finding evidence of non-exogeneity implies misspecification, but finding that some specific variables help predict certain shocks can shed light on the dimensions along which the model is misspecified. Forecast error variance decomposition analysis then helps assess the relevance of the missing channels. The paper puts the proposed methodology to work both in a controlled experiment - by running a Monte Carlo simulations with a known DGP - and using a state-of-the-art model and US data up to 2011.