I am trying to run a spatio-temporal autoregressive model (STAR). Therefore I need to create a spatial weight matrix W with N × T rows and N × T columns to weight country interdependencies based on yearly trade data. Could someone please tell me how I to create such a matrix in R or Stata?
Dear Jan, greetings from Paris, Montmartre !
I answer about the R case. First of all, it depends on your data, how they are initially organized. The pre-processing may involve many manipulations with tidyverse or data.table. Could you be more specific on your data sources ?
Finally, I seems very easy in R with the pivot_longer dplyr command but without some extra precisions it seems a bit hazy.
Concerning estimation there seems to have many packages :
https://arxiv.org/pdf/1812.01871.pdf
https://cran.r-project.org/web/packages/bmstdr/vignettes/bmstdr-vig_bookdown.html
https://cran.r-project.org/web/views/SpatioTemporal.html
https://spacetimewithr.org/code
All the best
Franck.
Dear Franck,
Thank you for your quick reply.
My data is based on country data per year. The main data set contains y x1... with variables in columns and countries and years in rows (please find an example attached). The data for the weights matrix is stored in a separate data set, where countries and years are also stored in the rows. However, the columns initially include trading data between country a and b in a given year (second attachment).
I hope this helps.
All the best to Paris from Heidelberg (Germany)
Jan
Dear Jan,
OK ! I see ! you need to create the spatial weight matrix indeed !
There are many possibilities in R:
https://cran.r-project.org/web/packages/spNetwork/vignettes/SpatialWeightMatrices.html
https://rpubs.com/erikaaldisa/spatialweights
I strongly advise to work with sf because it so easier now !
https://spatialanalysis.github.io/handsonspatialdata/applications-of-spatial-weights.html
but spdp may still be clearly adaptated to you context :
https://rdrr.io/rforge/spdep/man/nb2mat.html
This is one of the definitive books on the subject in R:
https://r-spatial.org/book/
with the book : Book Applied Spatial Data Analysis With R
there are other references but they are more geospatial (point process) oriented.
Here you should use one of those packages and that nb2mat from package spdp might do the trick !
All the best.
Franck.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Probability
- Probability Theory
- Reliability