search

nk027 / bvar

Toolkit for the estimation of hierarchical Bayesian vector autoregressions. Implements hierarchical prior selection for conjugate priors in the fashion of Giannone, Lenza & Primiceri (2015). Allows for the computation of impulse responses and forecasts and provides functionality for assessing results.
https://cran.r-project.org/package=BVAR
Other
50 stars 21 forks source link

Allow standardizing data #56

Open nk027 opened 4 years ago

nk027 commented 4 years ago

If data is standardized, the IRF need to be scaled up again. This should be supported somehow.

nk027 commented 4 years ago

Note that standardization is relevant for: https://github.com/nk027/bvar/blob/032d482640b44a7e550b3d85d95d80f76781df69/R/10_bvar.R#L266 https://github.com/nk027/bvar/blob/032d482640b44a7e550b3d85d95d80f76781df69/R/10_bvar.R#L267 https://github.com/nk027/bvar/blob/032d482640b44a7e550b3d85d95d80f76781df69/R/10_bvar.R#L268 https://github.com/nk027/bvar/blob/032d482640b44a7e550b3d85d95d80f76781df69/R/10_bvar.R#L269

howardya commented 4 years ago

Hello~

Do we know the effect of standardization to the accuracy of results?

Also, suppose I standardize the data manually, would anything break other than the IRF you mentioned above.

Thanks.

nk027 commented 4 years ago

Hey @howardya, The results should be fine and the IRF do not really break. It's just that they are then on a weird scale, which is tough to interpret.