I have a question about spdep package, ‘impacts’ function (in R). I’ve estimated spatial model (panel data). With this function I can calculate average direct and average indirect effects (averaged by region). But I don’t want them to be averaged. I want direct and indirect effects for the each region. What can I do?
If you're 100% it's an average, then you can just calculate it yourself by reversing the formula of the mean. Since and average is nothing more than the sum of values observated over the number of observation (in your case, the nº of regions), you can just multiply the average given by the function by the number of regions and you should have the effect itself. If you want to obtain a different value for each region that is not the average, then you should apply a grouping factor to your model. I suggest looking up the "glmm" package and the "lme4" package in R and see if it works for you.
Piras, G. (2014). Impact estimates for static spatial panel data models in R. Letters in Spatial and Resource Sciences, 7(3), 213–223. https://doi.org/10.1007/s12076-013-0113-8
- Badges
- Science topic
- Similar topics
- Mathematics
- Algebra
- Matrix