Detecting and Predicting Forecast Breakdowns

By R. Giacomini, Barbara Rossi. Social Science Research Network, 2009

202 citations recorded by the source corpus.

We propose a theoretical framework for assessing whether a forecast model estimated over one period can provide good forecasts over a subsequent period. We formalize this idea by defining a forecast breakdown as a situation in which the out-of-sample performance of the model, judged by some loss function, is significantly worse than its in-sample performance. Our framework, which is valid under general conditions, can be used not only to detect past forecast breakdowns but also to predict future ones. We show that main causes of forecast breakdowns are instabilities in the data generating process and relate the properties of our forecast breakdown test to those of existing structural break tests. The main differences are that our test is robust to the presence of unstable regressors and that it has greater power than previous tests to capture systematic forecast errors caused by recurring breaks that are ignored by the forecast model. As a by-product, we show that our results can be applied to forecast rationality tests and provide the appropriate asymptotic variance estimator that corrects the size distortions of previous forecast rationality tests. The empirical application finds evidence of a forecast breakdown in the Phillipsi curve forecasts of U.S. inflation, and links it to inflation volatility and to changes in the monetary policy reaction function of the Fed.

Subfields: Economics and Econometrics, Finance

View Paper: https://www.semanticscholar.org/paper/f5a4b27e4bdd3a47db8bc1cef4e3a4aecaf4f1cb

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

In Finance

Citations in the corpus

How Stable is the Forecasting Performance of the Yield Curve for Output Growth (2006, methodology): “As discussed in Giacomini and Rossi (2005), the t-test on the constant ai can be interpreted as a test of unconditional forecast breakdown (i.e. the tm,n,s forecast breakdown test).”
Advances in Forecasting Under Instability (2012, unknown): “53 (for simplicity, we assume that the same loss is used for both estimation and out-of-sample forecast evaluation) and forecast horizon h; Giacomini and Rossi (2009) propose analyzing the sequence of P out-of-sample surprise losses : SLt+h = Lt+h Lt, for t = R;R + 1; :::; T; (42) where Lt+h is the…”
Learning under Concept Drift: an Overview (2010, background): “Concept drift is relevant in making macroeconomic forecasts [58], predicting the phases of a business cycle [80].”
Testing for Structural Stability of Factor Augmented Forecasting Models (2014, methodology): “As shown in the (unnumbered) Proposition in Han and Inoue (2012), the criterion of Bai and Ng (2002), applied to the full sample, will detect 2r factors.”
Optimal forecasts in the presence of structural breaks (2011, background): “Contents lists available at ScienceDirect Journal of Econometrics journal homepage: www.elsevier.com/locate/jeconom Optimal forecasts in the presence of structural breaks✩ M. Hashem Pesaran a,b, Andreas Pick c,d,∗, Mikhail Pranovich e a University of Southern California, USA b Trinity College, Cambr”