Economics Asked on July 22, 2021
I want to fit a time-varying VAR model to my data because of a structural break occurring midway in the sample. But I want to plot IRFs, not for the entire dataset, but for the pre-structural-break period (about 30 data points) and the post-structural-break period (also about 30 data points). Does anybody know a package or software that can do this?
I have found packages that can do something similar in R (but for a time-varying BVAR). So this does not work for me as the packages require that I use the pre-structural break period (or part of it) as training data for estimating the prior coefficients. Thus, it does not allow one to impose their own priors for the coefficients (without estimating them using the data points at the beginning of the sample). Consequently, it estimates BVARs just for the remaining sample, which in my case will be mainly periods after the structural break.
My best alternative so far is to break the sample and build a VAR model for each period – before and after the structural break. Maybe a BVAR, given that the data is short for both periods.
Get help from others!
Recent Questions
Recent Answers
© 2024 TransWikia.com. All rights reserved. Sites we Love: PCI Database, UKBizDB, Menu Kuliner, Sharing RPP