Skip to content

Error with STFDF

4 messages · Philippi, Tom, Supadog, Benedikt Gräler

#
Hello,

I'm having trouble creating an STFDF object from my spacio-temporal data.
The object is intended to be used later to perform kriging interpolation. 
My data looks like this (a few lines):
      x    y   t            value
1    30   30 497826   6.6
2   960   30 499430   7.4
3  1890  30 500990   59.8

And I type in the following commands:
# read in the data from csv file and specify coordinates columns
f <- read.csv('data.csv') 
coordinates(f) <- ~x+y

# create SP object out of the coordinates
g <- as(f, "SpatialPoints")

# obtain time points. The timestamps are in milliseconds.
ts = f$t
options("digits.secs"=3)
time = as.POSIXct(ts/1e3, origin="1970-01-01", tz="Europe/Dublin")

# create data.frame
data = as.data.frame(f[2])

# finally create the STFDF object
obj = STFDF(g, time, data)

But I get the following error message:
Error: nrow(object at data) == length(object at sp) * nrow(object at time) is not
TRUE


What am I missing? The data is in long form so each time reading has its own
coords and data value and I don't want the equation in the error to hold. 



--
View this message in context: http://r-sig-geo.2731867.n2.nabble.com/Error-with-STFDF-tp7584461.html
Sent from the R-sig-geo mailing list archive at Nabble.com.
#
Hi Tom, 

Thanks for the response. I would like to interpolate for values in the
'value' column given different x, y and t values. I was hoping to use the
krigeST function but I just did a little bit of searching and it appears as
though it won't work with anything other than STFDF objects. Do I have any
other options?

Thanks,
Ziggy



--
View this message in context: http://r-sig-geo.2731867.n2.nabble.com/Error-with-STFDF-tp7584461p7584468.html
Sent from the R-sig-geo mailing list archive at Nabble.com.
#
Dear Ziggy,

the spatio-temporal sum-metric variogram model is the only one that is
currently also implemented for the STSDF classes. You should be able
to coerce your STIDF to a STSDF using as(). In the STSDF case ("S"
meaning sparse), the space and time instances are stored once and an
index relates observations to their spatio-temporal location. It can
handle both extreme cases of a full layout (ST-F-DF) and the irregular
case (ST-I-DF). See Figure 1 in the spacetime vignette [1] for a
graphical representation of all four ST* layouts.

Best,

 Ben

[1] http://cran.r-project.org/web/packages/spacetime/vignettes/jss816.pdf
On 27.08.2013 01:03, Supadog wrote:
Sent from the R-sig-geo mailing list archive at Nabble.com.