gwr.gauss

gwr.Gauss

The gwr.gauss function returns a vector of weights using
the Gaussian scheme:
$$w(g) = e^{{-(d/h)}^2}$$
where $d$ are the distances between the observations
and $h$ is the bandwidth.
The default (from release 0.5) gwr.Gauss function returns a vector of 
weights using the Gaussian scheme:
$$w(g) = e^{-(1/2) {{(d/h)}^2}}$$

spatial

Functions for computing geographically weighted
regressions are provided, based on work by Chris
Brunsdon, Martin Charlton and Stewart Fotheringham.

Roger Bivand

spgwr

Geographically Weighted Regression

Danlin Yu

Tomoki Nakaya

Miquel-Angel Garcia-Lopez

gwr.gauss function

<dl> <dt>dist2</dt>
<dd>vector of squared distances between observations and fit point</dd> <dt>bandwidth</dt>
<dd>bandwidth</dd></dl>

Arguments

Roger Bivand <a href="/link/Roger.Bivand%40nhh.no?package=spgwr&version=0.6-37" data-mini-rdoc="spgwr::Roger.Bivand@nhh.no">Roger.Bivand@nhh.no</a>

Author

The gwr.gauss function returns a vector of weights using
the Gaussian scheme:
$$w(g) = e^{{-(d/h)}^2}$$
where $d$ are the distances between the observations
and $h$ is the bandwidth.
The default (from release 0.5) gwr.Gauss function returns a vector of 
weights using the Gaussian scheme:
$$w(g) = e^{-(1/2) {{(d/h)}^2}}$$

GWR Gaussian weights function — gwr.gauss

<dl>

 <dt>dist2</dt>
<dd>vector of squared distances between observations and fit point</dd>

 <dt>bandwidth</dt>
<dd>bandwidth</dd>

</dl>

Roger Bivand <a href='mailto:Roger.Bivand@nhh.no'>Roger.Bivand@nhh.no</a>

gwr.gauss: GWR Gaussian weights function

Description

Usage

Value

Arguments

Author

References

See Also

Examples