Tailored randomized block MCMC methods with application to DSGE models

Chib, Siddhartha; Ramamurthy, Srikanth

doi:10.1016/j.jeconom.2009.08.003

Cited by 121 publications

(108 citation statements)

References 26 publications

Supporting

Mentioning

108

Contrasting

Order By: Relevance

“…We compare the algorithm using four different models: a small Real Business Cycle Model, the Generic State Space model of Chib and Ramamurthy (2010), the Smets and Wouters (2007) model and the model of Schmitt-Grohe and Uribe (2010). For each of the models, we calculate the "wall" (clock) time it takes to evaluate the likelihood at a particular point in the posterior 1000 times.…”

Section: Four Examplesmentioning

confidence: 99%

See 1 more Smart Citation

Using the “Chandrasekhar Recursions” for Likelihood Evaluation of DSGE Models

Herbst

2014

Comput Econ

View full text Add to dashboard Cite

In likelihood-based estimation of linearized Dynamic Stochastic General Equilibrium (DSGE) models, the evaluation of the Kalman Filter dominates the running time of the entire algorithm. In this paper, we revisit a set of simple recursions known as the "Chandrasekhar Recursions" developed by Morf (1974) and Morf, Sidhu, and Kalaith (1974) for evaluating the likelihood of a Linear Gaussian State Space System. We show that DSGE models are ideally suited for the use of these recursions, which work best when the number of states is much greater than the number of observables. In several examples, we show that there are substantial benefits to using the recursions, with likelihood evaluation up to five times faster. This gain is especially pronounced in light of the trivial implementation costs -no model modification is required. Moreover, the algorithm is complementary with other approaches.JEL Classification: C18, C63, E20

show abstract

Section: Four Examplesmentioning

confidence: 99%

“…The next example is the Generic State Space model used in Chib and Ramamurthy (2010). This is not a DSGE model.…”

Section: Generic State Space Modelmentioning

confidence: 99%

Using the “Chandrasekhar Recursions” for Likelihood Evaluation of DSGE Models

Herbst

2014

Comput Econ

View full text Add to dashboard Cite

show abstract

“…After obtaining the likelihood of the observables given the parameters, we use a Tailored Randomized Block Metropolis-Hastings (TaRB-MH) algorithm (Chib and Ramamurthy, 2010) to maximize the posterior likelihood. The prior distributions of the parameters, which are relatively weak, are given in Table 2.…”

mentioning

confidence: 99%

Policy risk and the business cycle

Born

Pfeifer

2014

Journal of Monetary Economics

378

253

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 AbstractThe argument that policy risk, i.e. uncertainty about monetary and fiscal policy, has been holding back the economic recovery in the U.S. during the Great Recession has a large popular appeal. We analyze the role of policy risk in explaining business cycle fluctuations by using an estimated New Keynesian model featuring policy risk as well as uncertainty about technology. We directly measure uncertainty from aggregate time series using Sequential Monte Carlo Methods. While we find considerable evidence of policy risk in the data, we show that the "pure uncertainty"-effect of policy risk is unlikely to play a major role in business cycle fluctuations. With the estimated model, output effects are relatively small due to i) dampening general equilibrium effects that imply a low amplification and ii) counteracting partial effects of uncertainty. Finally, we show that policy risk has effects that are an order of magnitude larger than the ones of uncertainty about aggregate TFP. JEL-Classification: E32, E63, C11

show abstract

“…Each minimum ( min ) is produced based on the features of the model (number of observations, number of endogenous variables, number of lags), and the optimal lambda ( b ) is calculated using the Markov Chain Monte Carlo with the Metropolis Hastings acceptance method (with 110,000 replications, we discard the …rst 10,000 ones and the following 100,000 replications are used in the estimation). Although we use the Random Walk -Metropolis Hastings (RW-MH) algorithm, we account for some problems it presents with the medium scale DSGE model of Smets-Wouters (2007) as shown in Chib and Ramamurthy (2010). In particular they propose replacing the commonly used single block RWM algorithm with a Metropolis-within-Gibbs algorithm that cycles over multiple, randomly selected blocks of parameters.…”

Section: Resultsmentioning

confidence: 99%

“…In particular they propose replacing the commonly used single block RWM algorithm with a Metropolis-within-Gibbs algorithm that cycles over multiple, randomly selected blocks of parameters. Chib and Ramamurthy (2010) provide evidence that the RW-MH algorithm has a serial correlation problem at lags 2500 and it is proven di¢ cult to tune up due to the dimensionality of the parameter space and the complexity of the posterior surface in case of many parameters, as with the Smets and Wouters (2007) model which estimates 36 parameters. To avoid the autocorrelation problem, Chib and Ramamurthy (2010) developed a Tailored randomized Block M-H (TaRB-MH) algorithm.…”

Section: Resultsmentioning

confidence: 99%

Bayesian forecasting with small and medium scale factor-augmented vector autoregressive DSGE models

Bekiros

Paccagnini

2014

Computational Statistics & Data Analysis

View full text Add to dashboard Cite

Publication information Computational Statistics and Data Analysis, 71 : 298-323Publisher Elsevier Item record/more information http://hdl.handle.net/10197/7322 Publisher's statementThis is the author's version of a work that was accepted for publication in Computational Statistics and Data Analysis. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in Computational Statistics and Data Analysis (VOL 71, ISSUE March 2014, (2014 AbstractAdvanced Bayesian methods are employed in estimating dynamic stochastic general equilibrium (DSGE) models. Although policymakers and practitioners are particularly interested in DSGE models, these are typically too stylized to be taken directly to the data and often yield weak prediction results. Hybrid models can deal with some of the DSGE model misspeci…cations. Major advances in Bayesian estimation methodology could allow these models to outperform well-known time series models and e¤ectively deal with more complex real-world problems as richer sources of data become available. A comparative evaluation of the out-of-sample predictive performance of many di¤erent speci…cations of estimated DSGE models and various classes of VAR models is performed, using datasets from the US economy. Simple and hybrid DSGE models are implemented, such as DSGE-VAR and Factor Augmented DSGEs and tested against standard, Bayesian and Factor Augmented VARs. Moreover, small scale models including the real gross domestic product, the harmonized consumer price index and the nominal short-term federal funds interest rate, are comparatively assessed against medium scale models featuring additionally sticky nominal prices, wage contracts, habit formation, variable capital utilization and investment adjustment costs. The investigated period spans 1960:Q4 to 2010:Q4 and forecasts are produced for the out-of-sample testing period 1997:Q1-2010:Q4. This comparative validation can be useful to monetary policy analysis and macro-forecasting with the use of advanced Bayesian methods.

show abstract

Tailored randomized block MCMC methods with application to DSGE models

Cited by 121 publications

References 26 publications

Using the “Chandrasekhar Recursions” for Likelihood Evaluation of DSGE Models

Using the “Chandrasekhar Recursions” for Likelihood Evaluation of DSGE Models

Policy risk and the business cycle

Bayesian forecasting with small and medium scale factor-augmented vector autoregressive DSGE models

Contact Info

Product

Resources

About