[R-sig-Geo] Fwd: spatio temporal variogram

[Hodgess, Erin]

Hi!

When you make your shift from the first data set to the second, set it up such the the X1 - X4 are time series.

For instance, if your data frame is called d1.df,
d1.df$X1 <- ts(d1.df$X1,start=1981,freq=12)
(for monthly data)

Hope this helps!

Sincerely,
Erin

________________________________________
From: r-sig-geo-bounces at r-project.org [r-sig-geo-bounces at r-project.org] on behalf of Jonesmus Wambua [jonesmus.mutua at gmail.com]
Sent: Wednesday, February 13, 2013 3:25 AM
To: r-sig-geo at r-project.org
Subject: [R-sig-Geo] Fwd: spatio temporal variogram

Hi everyone,
am trying to do temporal and cross correlation and also fit a
spatio-temporal variogram with data in this format.

 |           ID year   y  X_COORD  Y_COORD
        1   1  1981   60  36.98203  2.042057
        2   2  1982   40  37.12720  2.016312
        3   2  1983   20  37.12720  2.016312
        4   3  1980  165  35.38416  1.714100
        5   3  1981   26  35.38416  1.714100
        6   3  1982   50  35.38416  1.714100|

My challenge is getting the data in the right format. my years are only 5
but with over 500 cluster points. I have summarized the data by adding the
data for each cluster per year so that my data now looks like below

 year X1 X2 X3 X4
1 1980 NA NA  3 NA
2 1981  1 NA  2  5
3 1982 NA  0  0 NA
4 1983 NA  0 NA NA
5 1984 NA NA NA NA


where X1 are my clusters. when i tried to do a autocorrelation using the
following code,
acf(na.omit(som)) it works, but for cross-autocorrelation as follows

acf(na.omit(som), "xts") , it gives the following error

Error in ts(x) : 'ts' object must have one or more observations


can anyone help?
thanks

        [[alternative HTML version deleted]]

_______________________________________________
R-sig-Geo mailing list
R-sig-Geo at r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-geo