On detectable and non-detectable structural change

By David F. Hendry. 2000

116 citations recorded by the source corpus.

Abstract A range of parameter changes in I (1) cointegrated time series are not reflected in econometric models thereof, in that many shifts are not easily detected by conventional tests. The breaks in question are changes that leave the unconditional expectations of the I (0) components unaltered. Thus, dynamics, adjustment speeds etc. may alter without detection. However, shifts in long-run means are generally noticeable. Using the VECM model class, the paper discusses such results, explains why they occur, and uses Monte Carlo experiments to illustrate the contrasting ease of detection of ‘deterministic’ and ‘stochastic’ shifts.

Subfields: Economics and Econometrics

View Paper: https://www.semanticscholar.org/paper/d8880abe873a2124b998b8b33b12a7224d24dd87

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

An Automatic Test of Super Exogeneity (2010, background): “…[xT1+h] = ϕ ∗ − (Π∗)h (ϕ∗ −ϕ) = ϕ∗h −−−→h→∞ ϕ ∗ whereϕ∗ = (In −Π∗)−1 φ∗, so that, from (4): xT1+h −ϕ∗ = Π∗ (xT1+h−1 −ϕ∗) + νT1+h (5) Clements and Hendry (1994), Hendry and Doornik (1997), and Hendry (2000) show that changes inϕ are easy to detect, whereas those inφ andΠ are not whenϕ is unchanged.”
Pooling of Forecasts (2004, unknown): “This can be traced back at least to Rosenblatt (1952), with recent contributions by Shephard (1994), Kim et al. (1998) and Diebold et al.”
Robustifying forecasts from equilibrium-correction systems (2006, methodology): “All computations are based on PcGive: see Doornik and Hendry (2001). The first model was based on using the competitive interest ra te Rla, and the second on the opportunity-cost measure Rnet appropriate after the Banking Act of 1984 legalized interes t payments on chequing accounts.”
Economic Forecasting: Some Lessons from Recent Research (2003, background): “For example, even quite large shifts in the coefficients of mean-zero stochastic variables have small effects on forecast errors: see Hendry and Doornik (1997), Hendry (2000b) and section 5.6.”
Advances in Forecasting Under Instability (2012, unknown): “They note that structural breaks resulting in location (mean) shifts are one of the major causes of forecast failure, as discussed in Clements and Hendry (1998, 2002 and 2006), whereas shifts on variables that have mean zero have smaller impact on forecasts (Hendry, 2000).”
Non-Parametric Direct Multi-step Estimation for Forecasting Economic Processes (2005, methodology): “The fact that a modeller may be liable to omit such a variable as ηt justifies the use of DMS: as seen in Hendry (2000), location shifts often occur in macro-econometric analyses, but growth rate changes can be difficult to detect.”
On the prevalence of trends in primary commodity prices (2006, unknown): “Additionally and following Hendry (1999), the use of such critical values delivers the highest possible power for the test used, imperative considering the small sample size employed.”
Explaining Cointegration Analysis: Part II (2000, background): “Imposing valid cointegration restrictions can improve forecast accuracy somewhat (see Clements and Hendry (1995)), but for short-horizon forecasts, any potential benefit can be overwhelmed by changes in the parameters and (surprisingly, changes to the other parameters are less pernicious: see Hendr”
Predicting Markov-Switching Vector Autoregressive Processes (2000, background): “A theory of economic forecasting applicable to time series subject to deterministic breaks has been presented in Clements and Hendry (1999).”