Thank you Bastien, and Dr. Bivand!
I appreciate your suggestions.
Milan
On Friday, April 11, 2014 6:58 AM, Bastien Ferland-Raymond
wrote:
Hello,
I'm not familliar with SAR and CAR model, but smaller AIC is better than
larger AIC. The important thing here is to have the same Y vector (which
Hi all,
I fitted three models (Ordinary least squares, CAR, and SAR) as follows.
W<-nb2listw(nlist,style="B")
sar<-spautolm(Y~X,data=data.df,listw=W)
car<-spautolm(Y~X,listw=W,data=data.df, family="CAR")
ols<-lm(Y~X,data=data.df)
AIC(ols,car,sar)
df AIC
ols 3 -124.4118
car
Hi all,
I know there are several postings on this problem, but that did not help to
solve my problem
sarlm<-lagsarlm(Agr_2001~Gras_2001+For_2001,data=data1,listw=qw)
Error in solve.default(inf, tol = tol.solve) :
system is computationally singular: reciprocal condition number = 4.22828e-37
How
Hi all,
I read in a paper that has calculated the DIC (deviance information criteria)
value of SAR model.
Using spautolm (), I fitted SAR model, and DIC() does give an error.
> sar<-spautolm(Y~X1+X2,data=data,listw=listw)
> DIC(sar)
Error: could not find function "DIC"
How do I calculate DIC val
Hi all,
Just wondering why AIC of a predictive process model is not defined. Can any
body clarify?
Milan
[[alternative HTML version deleted]]
___
R-sig-Geo mailing list
R-sig-Geo@r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-geo
Hi Roger,
If I am using a number of packages from R, what is the general practice of
citation (to cite all the packages individually or to cite R as a whole) in
scientific journal publications?
Thanks,
Milan
___
R-sig-Geo mailing list
R-sig-Geo@r-proj
Hi Roger,
Please let me know if the following way of citing spgwr is incorect:
In references: Bivand, R., Yu, D.
(2013) spgwr: Geographically weighted
regression. R package version 3.0
and within the text:
We usedthe spgwr
package (Bivand and Yu 2013) for GWR in the statistical software pack
Hi all,
As I already mentioned, this is the problem due to one extra county
"yellowstone national" in montana. I am trying to plot Montana and wyoming
together for further analysis.. At the last line, index does not match between
counties$names and data$co(see NA below). I know this is is due t
Hi list,
when I run the code
map1 <- map("county",c("montana"), fill=T, plot=F)
I see that R has 57 counties stored for Montana, USA. It shows separate
counties "montana.yellowstone" and "montana.yellowstone national" in there.
However, in fact, Montana has only 56 counties. Has any on
Hi all,
I am familiar with fitting zip model in R for non spatial data. Can some one
direct me to some source where I can find such modeling in R for spatial data?
Thanks,
Milan
[[alternative HTML version deleted]]
___
R-sig-Geo mailing list
R-
Dear all,
I fitted a SAR model with a response and 3 predictors. The output has
coefficients for an intercept and the three predictors. However, summary of the
SAR model says that number of parameters estimated is 6. Can somebody clarify
why number of parameters estimated is 6, instead of 4 (inc
in
place of 1 standard error? or why not 0.5 standard error?
If you could clarify this to me, I would really appreciate that.
Milan Sharma
[[alternative HTML version deleted]]
___
R-sig-Geo mailing list
R-sig-Geo@r-project.org
https
Hi all,
as of now, I have been fitting mixed models for longitudinal data (measurements
repeated over time). I have spatial data (ID, response, predictors, longitude,
latitude). Generally mixed model in R uses something like : (1|Subject) in
the argument.
Now, my question is how do I take care
An output of data.frame form can be exported using "write.csv()"or
"write.table"
Milan
- Original Message -
From: "Hodgess, Erin"
To: Lilis Husna ; "r-sig-geo@r-project.org"
Cc:
Sent: Wednesday, February 27, 2013 11:23 PM
Subject: Re: [R-sig-Geo] save output to excel;
What kind
Hi list/ Roger,
I tried just 3 states as you said, I merged state and county names to make sure
the county names are different.
Here is the error message.
library(maptools)
library(maps)
library(spdep)
library(rgdal)
library(alr3)
data<-read.csv("try3.csv",header=TRUE)
> attach(data)
> names(dat
Dear List,
I fitted GWR, but different states have sane county names. Here is the code and
error message.
data<-read.csv("C:GWRDATA1.csv",header=TRUE)
> library(spgwr)
> map = SpatialPointsDataFrame(data=data,
> coords=cbind(data$Longitude,data$Latitude))
> SD.adapt.gauss <- gwr.sel(Y~X1+X2,data=
Dear list/ Dr. Bivand,
is it possible to plot spplot if there is no Longitude and Latitude in my data?
(i.e. asking R to find latitude and longitude itself, is there any way to do
that)?
Milan
___
R-sig-Geo mailing list
R-sig-Geo@r-project.org
https://
Dear all,
I fitted gwr and ols and I was expecting that residuals for gwr will be smaller
than that of ols. But, my result (copied below) is kind of mixed. Some ols
residuals are smaller than gwr, some are not.
Is it because of small samples? or anything else?
Milan
-0.136016648 1.10145715
Dear list,
I did GWR residual plot using the same technique as spgwr package. When I do
for Georgia data, I get the nice plot (with the state map) with the residuals,
but same code applied to my data gives residuals plot (without the state map).
My code is given below, if someone can tell me w
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, d
Dear list/ Professor Bivand,
I have a curiosity about spatial data. We have seen fitting CAR and
SAR models for county level data (real data). Is it possible to
simulate such a county level data and then perform such CAR and SAR
analysis for the simulated data?
I kind
Looks like you want autologistic regression.
Just google "autologistic modelling in R" and you will get archieves of
previous sessions.
Best,
Mitra Devkota
From: Tanja Zinßer
To: r-sig-geo@r-project.org
Sent: Wednesday, February 6, 2013 5:53 AM
Subject: [R-si
Dear list,
what command in R gives all the local parameter estimates of a GWR?
I just got minimum, max,of the coefficients.
Any help would be appreciated.
Milan
[[alternative HTML version deleted]]
___
R-sig-Geo mailing list
R-sig-Geo@r-proj
Dear list,
I have a data set, with response=yield of crop, predictors=temperature and
precipitation for all the counties of a state, repeatedly for many years. (i.e.
for each year, I have the data with these three variables for each country for
4 years).My question is, how can I take time into
Dear all,
I see that moran's correlogram can be constructed by correlog() and also by
sp.correlogram().
Do they have similar interpretations? Is there anything like these two should
be used in different conditions? or Can anyone of these be used in all the
conditions?
Thanks,
Milan
[[alt
Here is my problem in R:
> readOGR(dsn="C:/Users/abc/desktop/shapfile","SD06")
OGR data source with driver: ESRI Shapefile
Source: "C:/Users/abc/desktop/shapfile", layer: "SD06"
with 66 features and 36 fields
Feature type: wkbPolygon with 2 dimensions
#Is the above program running correctly? Or,
Dear all,
my question is about shapefile and GIS, rather than R. If somebody could kindly
help me, or can direct me to another site, I would appreciate that.
I created my shapefile successfully, and the name of my shapefile is AZ. I have
different name for the columns of my shape file, but shape
@ Chris: thanks! I just did spCbind to merge residuals with my spatial data
and the program worked just fine. Thanks :)
Now, how can I add site locations on the plot (like the one Bivand et. al.,
figure 10.5)?
Milan
From: Chris Fowler
Sent: Thursday, Ja
Dear all,Â
may be this one is readable.
I want to plot Poisson regression spatial residual plot, all the codes run but
there is a small error at the end in spplot command, which is shown below.
Can someone figure out what is wrong?
##
> GLMp <- glm(Tcrash~X100M
quot;
>
Milan
From: Virgilio Gómez-Rubio
Cc: "r-sig-geo@r-project.org"
Sent: Tuesday, January 8, 2013 1:42 PM
Subject: Re: [R-sig-Geo] Poisson regression spatial residual plot
Dear Milan,
El mar, 08-01-2013 a las 06:48 -0800, Milan Sharma escribió
Dear list,
I wanted to plot residuals from poisson regression model (like the figure 10.5
of Applied spatial data analysis with R, Bivand et. al.).
My data has response,Tcrash, and the covariates,
100MVMT, Alcohol PVMT Alignment
and longitude and latitude.
I used their code, but what "TCE"
Hi all,
can any one of you see whats wrong in the code below?
> model<-lm(Tcrash~X100MVMT+Alcohol+Alignment+PVMT, data=data)#works
> sar<-spautolm(Tcrash~X100MVMT+Alcohol+Alignment+PVMT,listw=WI_queenW,
> data=data)#works
> lag <- lagsarlm(Tcrash~X100MVMT+Alcohol+Alignment+PVMT, data=data,
> l
-sig-Geo] Shape file of USA land cover data
sessionInfo() will tell you what's loaded and install.packages("rgdal") in a
fresh session will install latest, depending on your system
On Thursday, December 6, 2012, Milan Sharma wrote:
Mike, I just updated my R (I uninstalled the previo
Hi List,
I wanted to download national land cover data of USA, 2006 which I could
convert to shape file and do analysis. The website
is http://www.mrlc.gov/nlcd01_data.php.
I got the following 6 files in it:
1. appendix1_nlcd2006_scene_list_by_path_row_txt
2. appendix1_nlcd2006_scene_list_by_pa
Dear all,
I have a shape file in my desktop. I could not read this file in R. The
following are my codes, can somebody tell me how to do? None of these are
working..
require(maptools)
library(rgdal)
##my working directory is "C:/Users/study/Desktop".
##name of my shape file is data.
shape <- re
Dear all,
I want to fit an exponential variogram for my lattice data. But, I just have
response variable (yield), and predictors: temperature and precipitation. I do
not have longitude and latitude.
This is the data of all the counties of a state of the USA. Is it possible to
fit exponential va
36 matches
Mail list logo