Reconciling output gaps: Unobserved components model and Hodrick–Prescott filter

Grant, Angelia L.; Chan, Joshua C. C.

doi:10.1016/j.jedc.2016.12.004

Cited by 49 publications

(36 citation statements)

References 35 publications

(57 reference statements)

Supporting

Mentioning

Contrasting

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

“…It is easy to see that if all the indicators are one, then the trend output y * t has a stochastic trend with a time-varying growth rate a 0 + a t . If δ a = 1 and δ y * = 0, then y * t is the same trend specification implicitly assumed by the Hodrick-Prescott filter (see, e.g, Grant and Chan, 2016b). If δ a = 0 and δ y * = 1, then y * t follows a random walk with drift δ a 0 a 0 .…”

Section: Trend Output Specification Searchmentioning

confidence: 95%

“…For instance, it is well known that the correlated unobserved components model of Morley, Nelson, and Zivot (2003) yields output gap estimates that are small in amplitude, whereas the Hodrick-Prescott (HP) filter generates large and persistent cycles. Grant and Chan (2016b) show that this difference arises entirely due to the different ways the trend output is modeled. In particular, the correlated unobserved components model assumes a random walk trend with a constant drift (or a small number of breaks in the drift), whereas the HP filter implicitly models the output trend as a particular random walk with a stochastic drift (see also Harvey and Jaeger, 1993).…”

Section: Trend Output Specification Searchmentioning

confidence: 98%

See 1 more Smart Citation

Measuring the Output Gap Using Stochastic Model Specification Search

Chan

Grant

2017

SSRN Journal

View full text Add to dashboard Cite

It is well known that different specification choices can give starkly different output gap estimates. To account for model uncertainty, we average estimates over a wide variety of popular specifications using stochastic model specification search. In particular, we consider three types of specification choices: sets of variables used in the analysis, output trend specifications and distributional assumptions. Using US data, we find that the unemployment gap is useful in estimating the output gap, but conditional on the unemployment gap, the inflation gap no longer depends on the output gap. Our results show a steady decline in trend output growth throughout the sample, and the estimate at the end of our sample is only about 1%. Moreover, data favor t over Gaussian distributed innovations, suggesting the relatively frequent occurrence of extreme events.

show abstract

“…Following a common practice in econometrics, we decompose the series of insulting epithets into a cyclical component and a trend. The cycle may be viewed as the demand side of the market (Baffigi et al 2013;Grant and Chan 2017), i.e. the extent to which the general population harbours a certain curiosity towards the topics related to homosexuality.…”

Section: Introductionmentioning

confidence: 99%

Language as a proxy for cultural change. A contrastive analysis for French and Italian lexicon on male homosexuality

Bevilacqua

Cerciello

2021

Qual Quant

View full text Add to dashboard Cite

Discrimination against LGBT people represents a significant and long-standing societal problem that occurs in several forms, including lexical discrimination, which consists in frequent usage of discriminatory epithets. Lexical discrimination produces a vicious circle where speakers grow subconsciously accustomed to abusive language and marginalisation becomes institutionalised. A vast literature has tackled lexical discrimination, providing several country-level studies. The cases of France and Italy are described as very different: while the French experience is centred around grassroot mobilisation, Italy features a traditional strategy of silence. This work aims to verify such difference empirically. Using a rich and detailed dataset, we apply time series analysis on the frequencies of usage of the terms that characterise male homosexuality. Our results highlight some similarities and some differences between the French and the Italian case, stressing the importance of lexical resemantisation that occurred in France but not in Italy.

show abstract

Reconciling output gaps: Unobserved components model and Hodrick–Prescott filter

Cited by 49 publications

References 35 publications

Large Hybrid Time-Varying Parameter VARs

Large Hybrid Time-Varying Parameter VARs

Measuring the Output Gap Using Stochastic Model Specification Search

Language as a proxy for cultural change. A contrastive analysis for French and Italian lexicon on male homosexuality

Contact Info

Product

Resources

About