Skip to contentScholarGate
LibraryBookshelfDeskReview StudioAssistant
Sign in
On this page
IntuitionHow it worksWhen to use itStrengths & limitationsCommon pitfallsApplicationsFrequently asked🔒 Read the full methodSourcesRelated methods
Cite this pageSpotted an issue on this page? Report or suggest a fix →
Home›Econometrics›Time-Varying Parameter VAR (TVP-VAR)
Regression modelMultivariate time series

Time-Varying Parameter VAR (TVP-VAR)

Also known as: Time-Varying Parameter Vector Autoregression, TVP-SVAR, Stochastic Coefficient VAR, Zamana Göre Değişen Parametreli VAR

TVP-VAR is a Bayesian multivariate time-series model in which both the VAR coefficients and the shock covariance matrix are allowed to evolve continuously over time as random walks. Introduced by Primiceri (2005) to study U.S. monetary policy transmission, the model captures structural changes and regime shifts without requiring ex-ante knowledge of when breaks occurred, making it indispensable for macroeconomics, finance, and any setting where economic relationships are suspected to be unstable across time.

ScholarGate
  1. Regression model
  2. v1
  3. 1 Sources
  4. PUBLISHED
Cite this page →
Tools & resources
Download slides
Learn & explore

Read the full method

Members only

Sign in with a free account to read this section.

Sign in

Method map

The neighbourhood of related methods — select a node to explore.

TVP-VAR
SVARVAR ModelTime-varying parameter V…

When to use it

Use TVP-VAR when theory or historical evidence suggests that the relationships among macroeconomic or financial variables shift over the sample period — for example, across monetary policy regimes, financial crises, or structural reforms. The model requires a moderately long time series (typically T ≥ 80 quarters) to identify time variation reliably. Key assumptions include random-walk parameter evolution and log-normally distributed stochastic volatility. It is less appropriate for short samples, high-dimensional systems without shrinkage priors, or purely cross-sectional questions. Alternatives include threshold VAR for discrete regime changes, factor-augmented VAR for large variable sets, or time-varying DFM for high-dimensional panels.

Strengths & limitations

Strengths
  • Simultaneously captures time variation in both coefficients and shock variances without pre-specifying break dates
  • Grounded in a coherent Bayesian framework that naturally quantifies parameter uncertainty via posterior credible intervals
  • Produces time-varying impulse response functions that reveal how policy or shock transmission evolves across regimes
  • Widely adopted benchmark in empirical macroeconomics, ensuring replicability and comparability with a large literature
Limitations
  • Computationally intensive MCMC estimation becomes prohibitive as the number of variables or lags grows (curse of dimensionality)
  • Requires long time series — sparse data make it difficult to distinguish genuine time variation from estimation noise
  • Random-walk assumption for parameters may be too restrictive when structural changes are abrupt rather than gradual
  • Posterior results can be sensitive to prior choices, particularly on innovation variances of the state equations

Frequently asked

How does TVP-VAR differ from a standard VAR with structural breaks?

A VAR with structural breaks assumes coefficients are piecewise constant and requires the researcher to specify — or statistically detect — exact break dates. TVP-VAR instead lets coefficients evolve continuously as random walks, so no breakpoints need to be fixed in advance. This flexibility comes at the cost of higher computational demand and a need for longer samples to identify smooth variation reliably.

Is stochastic volatility mandatory in a TVP-VAR?

No, but it is strongly recommended. Without stochastic volatility the model assumes constant error variances, which is implausible for macroeconomic data spanning different volatility regimes such as the Great Moderation or the Global Financial Crisis. Omitting it can cause the time-varying coefficients to absorb heteroskedasticity, biasing estimates of genuine parameter change. Primiceri (2005) includes stochastic volatility by default for this reason.

What software can estimate TVP-VAR?

Matlab code accompanying Primiceri (2005) is widely circulated and forms the basis of many extensions. R packages such as bvartools and BGVAR implement variants with convenient wrappers. Python users can leverage PyMC or Stan for custom specifications. Many central bank research departments maintain internal Fortran or Julia implementations optimised for speed when running large-scale TVP-VAR analyses.

Sources

  1. Primiceri, G. E. (2005). Time varying structural vector autoregressions and monetary policy. Review of Economic Studies, 72(3), 821–852. DOI: 10.1111/j.1467-937X.2005.00353.x ↗

How to cite this page

ScholarGate. (2026, June 2). Time-Varying Parameter VAR (TVP-VAR). ScholarGate. https://scholargate.app/en/econometrics/tvp-var

Related methods

SVARVAR Model

Which method?

Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.

  • SVAREconometrics↔ compare
  • VAR ModelEconometrics↔ compare
Compare side by side →

Referenced by

Time-varying parameter VECM

Similar methods

Time-varying parameter VAR modelTime-varying parameter SVAR modelTVP-FAVARTime-varying parameter AR modelTime-varying parameter Granger causalityTime-varying parameter VECMTime-varying parameter ARMA modelTime-varying parameter OLS

Related reference concepts

Bayesian NonparametricsHyperpriors and ShrinkageEconometric ModelingEconometricsGaussian Process ModelsHierarchical Bayesian Models

Spotted an issue on this page? Report or suggest a fix →

ScholarGate — TVP-VAR (Time-Varying Parameter VAR (TVP-VAR)). Retrieved 2026-07-21 from https://scholargate.app/en/econometrics/tvp-var · Dataset: https://doi.org/10.5281/zenodo.20539026
Quick facts
Originator
Giorgio Primiceri
Year
2005
Type
Bayesian state-space model
Subfamily
Multivariate time series
Estimation
Markov Chain Monte Carlo (MCMC)
Prior
Minnesota-type / diffuse priors on initial states
Related methods
SVARVAR Model
ScholarGate

A content-first reference library for research methods — what each one is, how it works, and where it comes from.

Open data (CC-BY)

Explore

  • Library
  • Search the library…
  • Browse by field
  • Fields
  • Journey
  • Compare
  • Which method?

Reference

  • Subjects
  • Atlas
  • Glossary
  • Methodology
  • Philosophy

Your tools

  • Bookshelf
  • Desk
  • Chat

Company

  • About
  • Pricing
  • Contact
  • Suggest a method

Entries are compiled from published sources for reference. Verifying the accuracy and suitability of any information for your own use remains your responsibility.

© 2026 ScholarGate · A research-method reference library
  • Privacy
  • Cookies
  • Terms
  • Delete account