(Un)Predictability and Macroeconomic Stability

By Antonello D’Agostino, Domenico Giannone, P. Surico. Social Science Research Network, 2006

185 citations recorded by the source corpus.

The ability of popular statistical methods, the Federal Reserve Greenbook and the Survey of Professional Forecasters to improve upon the forecasts of inflation and real activity from naive models has declined significantly during the most recent period of greater macroeconomic stability. The decline in the predictability of inflation is associated with a break down in the predictive power of real activity, especially in the housing sector. The decline in the predictability of real activity is associated with a break down in the predictive power of the term spread.

Subfields: Economics and Econometrics

View Paper: https://www.semanticscholar.org/paper/ea032ae0e85984b47dccdd6f071eaccc63dfc0cf

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Economics and Econometrics

Citations in the corpus

Nowcasting: the real time informational content of macroeconomic data releases (2008, background): “At horizons longer than the current quarter, there is very little forecastability for GDP (D’Agostino et al., 2006).”
Explaining the great moderation: it is not the shocks (2008, unknown): “As discussed by Forni et al. (2007), the shock can be recovered as the forecast error of a larger model, including those variables that may help forecasting GDP growth.”
Dynamic factor model with infinite‐dimensional factor space: Forecasting (2016, background): “Its impact on factor estimation and factor-based forecasts has been studied and discussed in Stock and Watson (2002a), D’Agostino et al. (2007), Banerjee et al. (2008), Stock and Watson (2009), D’Agostino et al. (2013), Clements (2015).”
VARMA versus VAR for Macroeconomic Forecasting (2008, background): “For example, D’Agostino et al. (2006) argued that multivariate models have not been as successful in forecasting GDP growth and inflation after the mid-1980s because the leading indicator properties of the spread have changed over this period.”
A comparison of forecast performance between federal reserve staff forecasts, simple reduced‐form models, and a DSGE model (2010, methodology): “However, according to Diebold and Mariano [1995] test, none of the reported poorer performance of the alternative forecasts reported for hours per capita in table 5 are significant in the statistical sense.”
Comparing Alternative Predictors Based on Large-Panel Factor Models (2012, unknown): “covariances of higher order helps, since it aligns the forecasts toward those of a random walk, which has a very good performance in this low persistent sample (see Atkeson and Ohanian, 2001; D’Agostino, Giannone, and Surico, 2006)”
Forecasting using a large number of predictors: Is Bayesian regression a valid alternative to principal components? (2008, unknown): “In the empirics we will use two algorithms recently proposed which work without limitations of dimensionality: LARS (Least Angle Regression) developed by Efron, Hastie, Johnstone, and Tibshirani (2004) and the Iterative Landweber scheme with soft thresholding at each iteration developed by De Mol an”