Testing for time variation in an unobserved components model for the U.S. economy

Berger, Tino; Everaert, Gerdie; Vierke, Hauke Hendrik

doi:10.1016/j.jedc.2016.05.017

Cited by 48 publications

(55 citation statements)

References 57 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The general evolution of the trend output growth is similar to those obtained previously in the literature (e.g. Berger, Everaert, and Vierke, 2016;Grant and Chan, 2017) using unobserved components models. Moreover, the timing of the apparent slowdown in trend output is consistent with the breakdates identified in Perron and Wada (2009) and Morley and Panovska (2019).…”

Section: Full Sample Resultssupporting

confidence: 85%

Large Hybrid Time-Varying Parameter VARs

Chan

2019

SSRN Journal

View full text Add to dashboard Cite

Time-varying parameter VARs with stochastic volatility are routinely used for structural analysis and forecasting in settings involving a few macroeconomic variables. Applying these models to high-dimensional datasets has proved to be challenging due to intensive computations and over-parameterization concerns. We develop an efficient Bayesian sparsification method for a class of models we call hybrid TVP-VARs -VARs with time-varying parameters in some equations but constant coefficients in others. Specifically, for each equation, the new method automatically decides (i) whether the VAR coefficients are constant or time-varying, and (ii) whether the error variance is constant or has a stochastic volatility specification. Using US datasets of various dimensions, we find evidence that the VAR coefficients and error variances in some, but not all, equations are time varying. These large hybrid TVP-VARs also forecast better than standard benchmarks. AbstractTime-varying parameter VARs with stochastic volatility are routinely used for structural analysis and forecasting in settings involving a few macroeconomic variables. Applying these models to high-dimensional datasets has proved to be challenging due to intensive computations and over-parameterization concerns. We develop an efficient Bayesian sparsification method for a class of models we call hybrid TVP-VARs-VARs with time-varying parameters in some equations but constant coefficients in others. Specifically, for each equation, the new method automatically decides (i) whether the VAR coefficients are constant or time-varying, and (ii) whether the error variance is constant or has a stochastic volatility specification. Using US datasets of various dimensions, we find evidence that the VAR coefficients and error variances in some, but not all, equations are time varying.These large hybrid TVP-VARs also forecast better than standard benchmarks.

show abstract

Section: Full Sample Resultssupporting

confidence: 85%

Large Hybrid Time-Varying Parameter VARs

Chan

2019

SSRN Journal

View full text Add to dashboard Cite

show abstract

“…First, Antolin-Diaz et al (2017) have recently shown that long-run growth in the U.S. is characterized by a slowly but persistently decreasing pattern. A similar result can be found in Berger et al (2016a). Second, the discussion whether persistence as measured by the AR coefficients has changed over time is more controversial.…”

Section: Results Stochastic Model Specification Searchsupporting

confidence: 71%

Measuring the international dimension of output volatility

Everaert

Iseringhausen

2018

Journal of International Money and Finance

Self Cite

View full text Add to dashboard Cite

This paper studies output fluctuations in a panel of OECD economies with the aim to decompose the evolution in output volatility into domestic and international factors. To this end we use a factor-augmented dynamic panel model with both domestic and international shocks and spillovers between countries through trade linkages. Changes in the volatility of output growth can be due to time-varying sensitivity to these shocks, changes in the propagation mechanism or shifts in the variances of shocks. We explicitly model cross-sectional dependence in the variance equation by specifying a common factor structure in the volatility of domestic shocks. The results show that while the size of international shocks and spillovers does not decrease in most countries, the volatilities of domestic shocks share a clear common decreasing trend. Hence, the 'Great Moderation' appears to be mainly driven by a decline in the volatility of domestic shocks rather than smaller international shocks. JEL classification: C32, E32, F44

show abstract

“…A persistent decline in potential output growth has recently been documented for the United States by Antolin-Diaz, Drechsel and Petrella (2017) and Berger, Everaert and Vierke (2016). This is in line with the literature on the US productivity slowdown in the early 1970s.…”

Section: An Unobserved Components Model With Stochastic Volatilitysupporting

confidence: 78%

“…Using quarterly data, Berger et al . () find much stronger evidence for time variation in the volatility of US output growth shocks. This is also confirmed by our Monte Carlo study in Appendix A.3 which suggests that indicator values can be low in small samples even in the presence of the non‐stationary component.…”

Section: Estimation Resultsmentioning

confidence: 92%

“…Deviations from the mean growth rate are assumed to be transitory that is the characteristic roots of the AR(2) process lie strictly outside the unit circle. The mean growth rate is assumed to evolve over time according to a random walk:

μ_{i, t} = μ_{i, t - 1} + η_{i, t}^{μ}, η_{i, t}^{μ} \overset{i . i . d .}{\sim} N false(0, σ_{μ, i}^{2} false)

A persistent decline in potential output growth has recently been documented for the United States by Antolin‐Diaz, Drechsel and Petrella () and Berger, Everaert and Vierke (). This is in line with the literature on the US productivity slowdown in the early 1970s.…”

Section: Empirical Modelmentioning

confidence: 99%

See 1 more Smart Citation

What Drives Output Volatility? The Role of Demographics and Government Size Revisited

Iseringhausen

Vierke

2018

Oxf Bull Econ Stat

Self Cite

View full text Add to dashboard Cite

This paper studies the determinants of output volatility in a panel of 22 OECD countries. In contrast to the existing literature, we avoid ad hoc estimates of volatility based on rolling windows, and we account for possible non‐stationarity. Specifically, output volatility is modelled within an unobserved components model where the volatility series is the outcome of both macroeconomic determinants and a latent integrated process. A Bayesian model selection approach tests for the presence of the non‐stationary component. The results point to demographics and government size as important determinants of macroeconomic (in)stability. A larger share of prime‐age workers is associated with lower output volatility, while higher public expenditure increases volatility.

show abstract

Testing for time variation in an unobserved components model for the U.S. economy

Cited by 48 publications

References 57 publications

Large Hybrid Time-Varying Parameter VARs

Large Hybrid Time-Varying Parameter VARs

Measuring the international dimension of output volatility

What Drives Output Volatility? The Role of Demographics and Government Size Revisited

Contact Info

Product

Resources

About