search

mboeck11 / BGVAR

Toolbox for the estimation of Bayesian Global Vector Autoregressions in R.
27 stars 20 forks source link

Weighting matrix, W.8016 #11

Closed bristowl closed 2 years ago

bristowl commented 2 years ago

The internal dataset W.8016 seems to be incorrect (to my knowledge).

When I run: rowSums(BGVAR::W.8016) I get:

image

I believe the rowSums of these weighting matrices should add up to one, as is the case with the other matrices in the package!

Session info: R version 4.0.3 (2020-10-10) Platform: x86_64-w64-mingw32/x64 (64-bit) Running under: Windows 10 x64 (build 19044)

Matrix products: default

locale: [1] LC_COLLATE=English_Australia.1252 LC_CTYPE=English_Australia.1252
[3] LC_MONETARY=English_Australia.1252 LC_NUMERIC=C
[5] LC_TIME=English_Australia.1252

attached base packages: [1] stats graphics grDevices utils datasets methods base

loaded via a namespace (and not attached): [1] Rcpp_1.0.8 lattice_0.20-41 zoo_1.8-9 assertthat_0.2.1
[5] MASS_7.3-53 cellranger_1.1.0 grid_4.0.3 R6_2.5.1
[9] bayesm_3.1-4 coda_0.19-4 RcppParallel_5.1.5 httr_1.4.2
[13] readxl_1.4.0 curl_4.3.2 GIGrvg_0.5 Matrix_1.2-18
[17] xts_0.12.1 keyring_1.3.0 tools_4.0.3 stochvol_3.2.0 [21] parallel_4.0.3 xfun_0.29 abind_1.4-5 compiler_4.0.3
[25] BGVAR_2.4.6 knitr_1.37

mboeck11 commented 2 years ago

Thanks for pointing this out, the weights have to row-standardized. Will be updated shortly.

bristowl commented 2 years ago

Thanks, a simple one!