Andrea Carriero

Real-Time Nowcasting with a Bayesian Mixed Frequency Model with Stochastic Volatility

Summary The paper develops a method for producing current quarter forecasts of gross domestic product growth with a (possibly large) range of available within‐the‐quarter monthly observations of economic indicators, such as employment and industrial production, and financial indicators, such as stock prices and interest rates. In light of existing evidence of time variation in the variances of shocks to gross domestic product, we consider versions of the model with both constant variances and stochastic volatility. We use Bayesian methods to estimate the model, to facilitate providing shrinkage on the (possibly large) set of model parameters and conveniently generate predictive densities. We provide results on the accuracy of nowcasts of realtime gross domestic product growth in the USA from 1985 through 2011. In terms of point forecasts, our proposal improves significantly on auto‐regressive models and performs comparably with survey forecasts. In addition, it provides reliable density forecasts, for which the stochastic volatility specification is quite useful.

Common Drifting Volatility in Large Bayesian VARs

The general pattern of estimated volatilities of macroeconomic and financial variables is often broadly similar. We propose two models in which conditional volatilities feature comovement and study them using U.S. macroeconomic data. The first model specifies the conditional volatilities as driven by a single common unobserved factor, plus an idiosyncratic component. We label this model BVAR with general factor stochastic volatility (BVAR-GFSV) and we show that the loss in terms of marginal likelihood from assuming a common factor for volatility is moderate. The second model, which we label BVAR with common stochastic volatility (BVAR-CSV), is a special case of the BVAR-GFSV in which the idiosyncratic component is eliminated and the loadings to the factor are set to 1 for all the conditional volatilities. Such restrictions permit a convenient Kronecker structure for the posterior variance of the VAR coefficients, which in turn permits estimating the model even with large datasets. While perhaps misspecified, the BVAR-CSV model is strongly supported by the data when compared against standard homoscedastic BVARs, and it can produce relatively good point and density forecasts by taking advantage of the information contained in large datasets.

Financial Factors, Macroeconomic Information and the Expectations Theory of the Term Structure of Interest Rates

In this Paper we concentrate on the hypothesis that the empirical rejections of the Expectations Theory (ET) of the term structure of interest rates can be caused by improper modeling of expectations. Our starting point is an interesting anomaly found by Campbell-Shiller (1987), when by taking a VAR approach they abandon limited information approach to test the ET, in which realized returns are taken as a proxy for expected returns. We use financial factors and macroeconomic information to construct a test of the theory based on simulating investors’ effort to use the model in ‘real time’ to forecast future monetary policy rates. Our findings suggest that the importance of fluctuations of risk premia in explaining the deviation from the ET is reduced when some forecasting model for short-term rates is adopted and a proper evaluation of uncertainty associated to policy rates forecast is considered.

Have standard VARs remained stable since the crisis

Small or medium-scale VARs are commonly used in applied macroeconomics for forecasting and evaluating the shock transmission mechanism. This requires the VAR parameters to be stable over the evaluation and forecast sample, or to explicitly consider parameter time variation. The earlier literature focused on whether there were sizable parameter changes in the early 1980s, in either the conditional mean or variance parameters, and in the subsequent period until the beginning of the new century. In this paper we conduct a similar analysis but focus on the effects of the recent crisis. Using a range of techniques, we provide substantial evidence against parameter stability. The evolution of the unemployment rate seems particularly different relative to its past behavior. We then discuss and evaluate alternative methods to handle parameter instability in a forecasting context.

Forecasting The Yield Curve Using Priors From No‐Arbitrage Affine Term Structure Models

I propose a strategy for forecasting the term structure of interest rates that may produce significant gains in predictive accuracy. The key idea is to use the restrictions implied by Gaussian, no-arbitrage, affine term structure models on a vector autoregression as prior information instead of imposing the restrictions dogmatically. This allows us to account for possible model misspecification. We use the proposed method to forecast a system of five U.S. yields up to 12 months ahead, and we find it provides significant gains in forecast accuracy.

Sectoral Survey-based Confidence Indicators for Europe

We analyse a novel dataset of Business and Consumer Surveys, using dynamic factor techniques, to produce composite coincident indices (CCIs) at the sectoral level for the European countries and for Europe. Surveys are timely available, not subject to revision, and fully comparable across countries. Moreover, the substantial discrepancies in activity at the sectoral level justify the interest in sectoral disaggregation. Compared with the confidence indicators produced by the European Commission we show that factor-based CCIs, using survey answers at a more disaggregate level, produce higher correlation with the reference series for the majority of sectors and countries.

Explaining US-UK Interest Rate Differentials: A Reassessment of the Uncovered Interest Rate Parity in a Bayesian Framework

In this paper, we test the uncovered interest rate parity (UIRP), allowing for transitory deviations from it. These deviations may arise from variations in risk premia, errors in expectations and linearization errors, and are modelled as a zero-mean noise around the restrictions implied by the UIRP on a vector autoregression (VAR) in the interest rate differential and the spot exchange rate. Importantly, this approach includes the traditional one as a special case, which is derived by simply setting the noise to zero. When the noise is set to zero the UIRP is rejected, but if we allow for some degree of noise the UIRP is strongly supported by the data. Thus the UIRP relation does not hold exactly, but on average, with a stationary risk premium as opposed to a constant one. This result implies that analysing the effects of policy experiments under the null of the UIRP may be both safe and useful.

UK Term Structure Decompositions at the Zero Lower Bound

This paper employs a Zero Lower Bound (ZLB) consistent shadow-rate model to decompose UK nominal yields into expectation and term premium components. Compared to a standard affine term structure model, it performs relatively better in a ZLB setting by capturing the stylized facts of the yield curve. The ZLB model is then exploited to estimate inflation expectations and risk premiums. This entails jointly pricing and decomposing nominal and real UK yields. We find evidence that medium- and long-term inflation expectations are contained within narrower bounds since the early 1990s, suggesting monetary policy credibility improved after the introduction of inflation targeting.

No-Arbitrage Restrictions and Yield Curve Forecasting

This paper proposes to use No-Arbitrage Affine Term Structure Models as prior information on a Vector Autoregression (VAR) of yields. We evaluate the forecasting performance of the proposed approach against alternative models such as an unrestricted VAR and a Random Walk. As a result, we show that using No-Arbitrage restrictions leads to significant improvements in forecasting the yield curve.