Description The fit functions The predrict function
The package provides two categories of important functions: fit and predict.
The fit functions gwlmm
and rgwlmm
fit a geographically weighted
linear mixed model (GWLMM) to data. The GWLMM is a nested error regression model that takes
into account spatial non-stationarity. The function gwlmm
assumes normality
for the error term components. This assumption can be violated in the presence
of outliers. The function rgwlmm
fits an outlier robust version of the GWLMM to the data.
The predict functions predict.gwlmm
and predict.rgwlmm
estimate
small area means based on the model fits from the fit funtions when population data is provided.
As predcision measures MSE estiamtes are provided.
The predict functions can handle aggregated population information with centroid coordinates and
unit-level population data with coordinates for each unit.
If no data is provided these functions provide the in-sample predictions
(random effects, residuals, etc.) based on the model fits.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.