spgwr (georgia)
On Sat, 9 Feb 2013, Roger Bivand wrote:
YOU ARE STILL POSTING HTML! Please do follow the rules, which are quite clear.
To clarify the above, this is a setting in email clients for outgoing email: http://www.r-project.org/mail.html#instructions General Instructions Note that you should configure your e-mail software in such a way as to send only plain text, i.e., no HTML. ?html-ified? messages are usually considerably longer (in bytes!) and harder to filter for spam or viruses. Many of these (e.g. ?html-only? ones) are currently spam-filtered or otherwise intercepted completely and without notice to the sender. The link in the instructions to http://expita.com/nomime.html is dead, this is a copy: http://stagecraft.theprices.net/nomime.html Another link is: http://ask-leo.com/whats_the_difference_between_plain_text_and_html_email.html but has lots of adverts. This is also helpful: http://email.about.com/od/netiquettetips/qt/When-In-Doubt-Send-Plain-Text-Email-Not-Fancy-Html.htm Roger (list admin.)
The object you are interested in is res.adpt$SDF,
so in R (as is easy to establish by reading the introductory manual) you can
say:
summary(res.adpt$SDF)
which tells you what it is, and what is inside it. Using str() fives more
details on the contents.
You are plotting the local coefficients for the PctBlack covariate, the
residuals are gwr.e. So:
res.adpt$SDF$ols.e <- residuals(lm(PctBach ~ TotPop90 + PctRural +
PctEld + PctFB + PctPov + PctBlack, data=gSRDF))
spplot(res.adpt$SDF, c("ols.e", "gwr.e"))
plots both together on the same scale.
Roger
On Sat, 9 Feb 2013, Milan Sharma wrote:
Dear all, #the following is gwr code for georgia data. #Can somebody point out what is plotted by this code? #I want to plot residual plot (gwr and ols both)?for this data. library(spgwr) data(georgia) g.adapt.gauss <- gwr.sel(PctBach ~ TotPop90 + PctRural + PctEld + PctFB + PctPov + PctBlack, data=gSRDF, adapt=TRUE) res.adpt <- gwr(PctBach ~ TotPop90 + PctRural + PctEld + PctFB + PctPov + PctBlack, data=gSRDF, adapt=g.adapt.gauss) brks <- c(-0.25, 0, 0.01, 0.025, 0.075) cols <- grey(5:2/6) plot(res.adpt$SDF, col=cols[findInterval(res.adpt$SDF$PctBlack, brks, all.inside=TRUE)]) #I want residual plot under GWR and OLS. [[alternative HTML version deleted]]
Roger Bivand Department of Economics, NHH Norwegian School of Economics, Helleveien 30, N-5045 Bergen, Norway. voice: +47 55 95 93 55; fax +47 55 95 95 43 e-mail: Roger.Bivand at nhh.no