VAR Forecasting Using Bayesian Variable Selection

Korobilis, Dimitris

doi:10.2139/ssrn.1564378

Cited by 37 publications

(63 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The resulting predictive equations closely resemble those obtained via Bayesian variable selection using discrete mixture prior. Korobilis (2013b) showed that, in a forecasting context, sparsity was highly competitive with traditional shrinkage methods. We saw in an empirical study that the predictive performance of the horseshoe prior is very close to that of the discrete mixture priors, and often beats it.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Achieving parsimony in Bayesian vector autoregressions with the horseshoe prior

Follett

2019

Econometrics and Statistics

View full text Add to dashboard Cite

Section: Discussionmentioning

confidence: 99%

“…ijk y j,t−k ). Korobilis (2013b) notes that for the discrete mixture prior, this is Bayesian model averaging.…”

Section: Predictive Performancesmentioning

confidence: 99%

Achieving parsimony in Bayesian vector autoregressions with the horseshoe prior

Follett

2019

Econometrics and Statistics

View full text Add to dashboard Cite

“…where − indexes all the elements of the vector = ( 1 , … , ) but the th, and the conjugate prior for each is thus the independent Bernoulli density. Exact expressions for the conditional densities can be found in Korobilis (2011). Here, we provide a pseudo-algorithm that shows that the algorithm for the restricted model only adds one block that samples the vector , in the standard algorithm of the unrestricted regression model:…”

Section: Model Selection and Regularizationmentioning

confidence: 99%

“…In particular, we use Stochastic Search Variable Selection (SSVS) priors (George & McCulloch, 1993) that favor shrinkage of the model parameters in an automatic fashion and require only minimal prior input from the researcher. As discussed in Korobilis (2011), sparsity in our model parameters can be induced at a minimal computational cost with small adjustments of the standard Gibbs sampler algorithm commonly used for Bayesian linear regressions (Koop, 2003). We discuss our model within a study in which we wish to characterize the shape changes of human mandible profiles over years.…”

Section: Introductionmentioning

confidence: 99%

Predictive functional ANOVA models for longitudinal analysis of mandibular shape changes

2019

View full text Add to dashboard Cite

In this paper, we introduce a Bayesian statistical model for the analysis of functional data observed at several time points. Examples of such data include the Michigan growth study where we wish to characterize the shape changes of human mandible profiles. The form of the mandible is often used by clinicians as an aid in predicting the mandibular growth. However, whereas many studies have demonstrated the changes in size that may occur during the period of pubertal growth spurt, shape changes have been less well investigated. Considering a group of subjects presenting normal occlusion, in this paper we thus describe a Bayesian functional ANOVA model that provides information about where and when the shape changes of the mandible occur during different stages of development. The model is developed by defining the notion of predictive process models for Gaussian process (GP) distributions used as priors over the random functional effects. We show that the predictive approach is computationally appealing and that it is useful to analyze multivariate functional data with unequally spaced observations that differ among subjects and times. Graphical posterior summaries show that our model is able to provide a biological interpretation of the morphometric findings and that they comprehensively describe the shape changes of the human mandible profiles. Compared with classical cephalometric analysis, this paper represents a significant methodological advance for the study of mandibular shape changes in two dimensions.

show abstract

“…Caner and Zhang, 2014), or macro-economics (e.g. Korobilis, 2013). A sparse cointegration approach is useful for several reasons.…”

Section: Introductionmentioning

confidence: 99%

Sparse Cointegration

Wilms

Croux

2014

SSRN Journal

View full text Add to dashboard Cite

Cointegration analysis is used to estimate the long-run equilibrium relations between several time series. The coefficients of these long-run equilibrium relations are the cointegrating vectors. In this paper, we provide a sparse estimator of the cointegrating vectors. The estimation technique is sparse in the sense that some elements of the cointegrating vectors will be estimated as zero. For this purpose, we combine a penalized estimation procedure for vector autoregressive models with sparse reduced rank regression. The sparse cointegration procedure achieves a higher estimation accuracy than the traditional Johansen cointegration approach in settings where the true cointegrating vectors have a sparse structure, and/or when the sample size is low compared to the number of time series. We also discuss a criterion to determine the cointegration rank and we illustrate its good performance in several simulation settings. In a first empirical application we investigate whether the expectations hypothesis of the term structure of interest rates, implying sparse cointegrating vectors, holds in practice. In a second empirical application we show that forecast performance in high-dimensional systems can be improved by sparsely estimating the cointegration relations.

show abstract

VAR Forecasting Using Bayesian Variable Selection

Cited by 37 publications

References 46 publications

Achieving parsimony in Bayesian vector autoregressions with the horseshoe prior

Achieving parsimony in Bayesian vector autoregressions with the horseshoe prior

Predictive functional ANOVA models for longitudinal analysis of mandibular shape changes

Sparse Cointegration

Contact Info

Product

Resources

About