The R/TVSR package is an R package designed to estimate coefficients in a time-varying spatial regression model. These estimates are based on the Local-Kernel Spatial Regression (LKSR) and Score-Driven Spatial Regression (SDSR) estimation procedures.
You can access the code on the Github page. I am currently working on making it available on CRAN.
R/TVSR includes two main functions to estimate time-varying spatial regression models:
1. Local-Kernel Spatial Regression (LKSR): This is a semi-parametric estimation procedure based on the local maximization of a log-likelihood function.
2. Score-Driven Spatial Regression (SDSR): This function implements a score-driven approach, based on the work of Blasquez (2016).
These programs were initially developed to assess international monetary spillovers between 1980 and 2020.
For more details, refer to: Peeters, Girard, and Gnabo, "Monetary Response or Economic Integration: What Drives International Monetary Policy Spillovers?" (forthcoming). More information on the paper can be found here.