Bayesian inference in a time varying cointegration model

Koop, Gary; Leon-Gonzalez, Roberto; Strachan, Rodney W.

doi:10.1016/j.jeconom.2011.07.007

Cited by 50 publications

(33 citation statements)

References 42 publications

(31 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These papers suggest there is considerable variation in the parameters, particularly over the 1970s. However, the reduced rank restrictions due to cointegration introduce further conceptual and computational issues for time‐varying models, and a first example of how to implement such a time‐varying VECM is presented in Koop et al (2011). Alternatively, in using a SVAR for business cycle analysis, one may use prior information on the length and amplitude of the period of oscillation (see Harvey et al 2007).…”

Section: Discussionmentioning

confidence: 99%

Evidence on Features of a Dsge Business Cycle Model From Bayesian Model Averaging*

Strachan

Dijk

2013

Int Economic Review

Self Cite

View full text Add to dashboard Cite

The empirical support for features of a Dynamic Stochastic General Equilibrium model with two technology shocks is evaluated using Bayesian model averaging over vector autoregressions. The model features include equilibria, restrictions on long‐run responses, a structural break of unknown date, and a range of lags and deterministic processes. We find support for a number of features implied by the economic model, and the evidence suggests a break in the entire model structure around 1984, after which technology shocks appear to account for all stochastic trends. Business cycle volatility seems more due to investment‐specific technology shocks than neutral technology shocks.

show abstract

Section: Discussionmentioning

confidence: 99%

Evidence on Features of a Dsge Business Cycle Model From Bayesian Model Averaging*

Strachan

Dijk

2013

Int Economic Review

Self Cite

View full text Add to dashboard Cite

show abstract

“…One would obtain the same likelihood by replacing the aforementioned products with

α V V^{- 1} β^{normalT}

and HUU −1 x t for any non‐singular matrices V and U of appropriate dimensions. The same holds for the marginal likelihood when V and U are orthogonal, and x in model (3)–(4) is integrated out, i.e.

p false(y false| α, β, H, B, Q, R, ξ false) = p false(y false| α V^{- 1}, V β, H U^{- 1}, U B U^{- 1}, U Q U^{T}, R, ξ false) .

This is a well‐known observation in the literature related to both cointegration (Koop et al ., ; Villani, ; Larsson and Villani, ) and factor models (Aßmann et al ., ; Bai and Wang, ; Jackson et al ., ; Chan et al ., ). The issue is akin to the notorious label switching problem in mixture models, e.g.…”

Section: A Cointegrated Vector Auto‐regression Model With Dynamic Facmentioning

confidence: 99%

“…Recent advances in the literature, such as Villani (), Larsson and Villani () and Koop et al . (), developed ways of eliciting priors over cointegration spaces (as opposed to directly defining a prior over the Euclidean parameter space defined by elements of matrix β ) and deriving simulation methods for the corresponding posteriors.…”

Section: Bayesian Inferencementioning

confidence: 99%

Factor-Augmented Bayesian Cointegration Models: A Case-Study on The Soybean Crush Spread

Marówka

Peters

Kantas

et al. 2020

Journal of the Royal Statistical Society Series C: Applied Statistics

View full text Add to dashboard Cite

Summary We investigate how vector auto‐regressive models can be used to study the soybean crush spread. By crush spread we mean a time series marking the difference between a weighted combination of the value of soymeal and soyoil to the value of the original soybeans. Commodity industry practitioners often use fixed prescribed values for these weights, which do not take into account any time‐varying effects or any financial‐market‐based dynamic features that can be discerned from futures price data. We address this issue by proposing an appropriate time series model with cointegration. Our model consists of an extension of a particular vector auto‐regressive model that is used widely in econometrics. Our extensions are inspired by the problem at hand and allow for a time‐varying covariance structure and a time‐varying intercept to account for seasonality. To perform Bayesian inference we design an efficient Markov chain Monte Carlo algorithm, which is based on the approach of Koop and his co‐workerss. Our investigations on prices obtained from futures contracts data confirmed that the added features in our model are useful in reliable statistical determination of the crush spread. Although the interest here is on the soybean crush spread, our approach is applicable also to other tradable spreads such as oil and energy‐based crack and spark spreads.

show abstract

“…This would be possible by using the timevarying VECM framework, by Bierens and Martins (2010), which is an extension of the methods proposed by Johansen (1988Johansen ( , 1991Johansen ( , 1995 and allows to analyze the development of the cointegration relationship over time. Another promising approach in this research area might be using a Bayesian framework, for example Koop et al (2011) also offer the possibility to explicitly allow the cointegration space to evolve over time. …”

Section: Conclusion and Further Researchmentioning

confidence: 99%

Do They Still Matter? – Impact of Fossil Fuels on Electricity Prices in the Light of Increased Renewable Generation

Lips

2017

Journal of Time Series Econometrics

View full text Add to dashboard Cite

Standard-Nutzungsbedingungen:Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden.Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen.Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in AbstractDuring the last years, the German energy sector and especially its electricity market was affected by a major energy transition, the so called "Energiewende". This transition led to an increase of electricity production from renewable sources and thereby affected the whole electricity market. Therefore, it provides lessons for countries, which are only beginning a similar transition away from fossil fuels to renewable energy sources. The aim of this analysis is to assess if there still exists a relationship between fossil fuel and electricity prices. Due to possible structural breaks in the time series a minimum Lagrange Multiplier (LM) stationarity test is applied, which endogenously determines possible structural breaks. Subsequently a bootstrap approach is used to estimate confidence intervals (C.I.s) for the test statistic and the possible break dates. Furthermore, the stability of the cointegration vector is assessed with the test by Hansen and Johansen (1999). The results indicate that the cointegration relationship is not stable over time. To incorporate these findings, the cointegration analysis is based on Johansen et al. (2000), which allows structural breaks in the deterministic part of the cointegration relation. These results supports the assumption that the energy transition affected the relationship between fossil fuels and electricity prices, although there still exists a relatively strong cointegration relation between fossil fuel and electricity prices in the long run.Keywords: Structural Breaks, Bootstrap, Stationarity Test, Cointegration, Energy Economics, German Energy Transition JEL classification: C58, G14, L94, Q41. * I am grateful to Peter Winker, all my colleagues and the participants of the MAGKS Colloquium (Marburg) for helpful comments and fruitful discussions.

show abstract

Bayesian inference in a time varying cointegration model

Cited by 50 publications

References 42 publications

Evidence on Features of a Dsge Business Cycle Model From Bayesian Model Averaging*

Evidence on Features of a Dsge Business Cycle Model From Bayesian Model Averaging*

Factor-Augmented Bayesian Cointegration Models: A Case-Study on The Soybean Crush Spread

Do They Still Matter? – Impact of Fossil Fuels on Electricity Prices in the Light of Increased Renewable Generation

Contact Info

Product

Resources

About