R and Stata Options for VAR–GARCH and BEKK Models
Summary
The discussion points to software options for estimating multivariate volatility models. One answer recommends RATS, describing built-in VAR-GARCH support with several covariance specifications, including CCC, DCC, VECH, and BEKK. Another suggests the MTS package in R for a BEKK(1,1) model, if that is what the questioner means by VAR-EGARCH.
The guidance is brief and does not establish that either option implements every possible VAR-EGARCH formulation. It gives no Stata-specific formula, comparison of packages, estimation details, or empirical results. The main practical lesson is to clarify the intended model specification, since the label VAR-EGARCH may refer to distinct multivariate GARCH structures, and then check that a candidate tool supports the required form. The suggestions are pointers rather than a full implementation guide, and the exchange offers no evidence about model performance or suitability for a particular thesis.
Key ideas
- RATS is suggested for VAR-GARCH estimation with several covariance structures.
- The MTS package in R is suggested for a BEKK(1,1) model.
- The answers do not provide a Stata implementation or explain how to estimate the models.
- The appropriate software depends on what the researcher means by VAR-EGARCH.
Tags
Full text
# Is there an implementation of VAR-EGARCH model in R or Stata? # Is there an implementation of VAR-EGARCH model in R or Stata? I am writing my undergrad honor thesis and want to run a multivariable VAR-EGARCH model. Is there any package in R or formulas in Stata 14 that allows me to implement directly? If not, could you give me some advice? ## Answer by Sadhak (score 1) https://quant.stackexchange.com/a/22804 You can use RATS software in which VAR GARCH is inbuilt function with CCC, DCC VECH and BEKK for co-variance estimation. ## Answer by Alejandro Andrade (score 0) https://quant.stackexchange.com/a/22171 If for VAR-EGARCH you mean a BEKK model the `MTS`package in R can implement a BEKK(1,1)
Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.