The spflow package allows to estimate spatial econometric interaction models. It is designed to exploit the relational structure of flow data, reducing the computational burden and memory requirements.
You can install development version from GitHub with:
# install.packages("devtools")
devtools::install_github("LukeCe/spflow")
The package provides a new sp_multi_network-class
that combines
information on the origins, the destinations, and the origin-destination
pairs. Once this object is created, we can estimate an interaction model
with the spflow()
function. The package
vignette
contains a more detailed example.
library("spflow")
spflow(y9 ~ . + G_(DISTANCE), multi_net_usa_ge)
#> --------------------------------------------------
#> Spatial interaction model estimated by: MLE
#> Autocorrelation structure: model_9 (SDM)
#> Observations: 256
#>
#> --------------------------------------------------
#> Coefficients:
#> est sd t.stat p.value
#> rho_d 0.48 0.03 15.75 0.02
#> rho_o 0.36 0.03 10.59 0.03
#> rho_w -0.25 0.04 -5.95 0.05
#> (Intercept) 10.00 1.92 5.21 0.06
#> (Intra) 11.32 3.07 3.69 0.08
#> DEST_X 0.94 0.06 14.97 0.02
#> DEST_X.lag1 0.62 0.11 5.80 0.05
#> ORIG_X -0.78 0.04 -21.20 0.02
#> ORIG_X.lag1 -0.32 0.08 -3.93 0.08
#> INTRA_X 1.95 0.08 23.98 0.01
#> INTRA_X.lag1 -0.34 0.19 -1.76 0.16
#> DISTANCE -2.80 0.34 -8.25 0.04
#>
#> --------------------------------------------------
#> R2_corr: 0.9927709
Any scripts or data that you put into this service are public.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.