R help - Oct 2011 - Randomized Points in space/ saving model results

A bit new to R and I'm working on something a bit more challenging than I am
used to- so here's whats going on:

Data inputs:  9 different shapefiles (.shp) of different point locations
(lat, long)
                   shapefile bounding box  (lat/long corner points (14, 1)
(15,1) (14, 2) (15,2))    
                   1 .csv of lat/long coordinates for points
                   


Goal:  a- I want to randomly drop x number of points in a study area within
the bounding box
         b- Then I want to add the .csv points to the mix
         c- Measure the distances from all these points to each of the 9
shapefiles locations, one at a time and save the results in a table or list
         d- run an regression analysis on the table or list created (I have
this code all set to go)
         e- I want to save the outputs of the model in a new table  (I have
this code also)
         f-  Run this whole process again x number of times and save x
number of outputs in the table in order of creation

Any help on steps a, b,c, or f would be appreciated

M.
         

--
View this message in context:
http://r.789695.n4.nabble.com/Randomized-Points-in-space-saving-model-results-tp3922612p3922612.html
Sent from the R help mailing list archive at Nabble.com.

Not doing much work with spacial stats or shapefiles, I can't help in
too much detail, but here are some R commands that might help for each
part:

a.

# This will help you pick random points within your bounded box
runif2d <- function(n, xmin, xmax, ymin, ymax){
     stopifnot(all(xmax > xmin, ymax > min))
     y <- runif(n, ymin, ymax)
     x <- runif(n, xmin, xmax)
     cbind(x,y)
}

b. take the output of runif2d (which gives a nx2 matrix) and cbind()
the csv points on as well

c. getdist <- function(inpoints, refpoints) {
    # Takes in two 2 column matrices representing (x,y) coordinates
and returns a matrix with all the distance pairs between them
       stopifnot(all(is.matrix(inpoints), is.matrix(refpoints),
dim(inpoints)[2L] == 2L, dim(refpoints)[2L] == 2L))
       d <- matrix(nrow = nrow(inpoints), ncol = nrow(refpoints))
       for (i in seq_along(refpoints)) {
             d[, i] <- rowSums(inpoints-refpoints[i,]^2)
       }
       d
}

f. Wrap everything in a function and use replicate()

Do these help?

Michael
On Thu, Oct 20, 2011 at 1:32 PM, magono <nroyal01 at gmail.com>
wrote:
> A bit new to R and I'm working on something a bit more challenging than
I am
> used to- so here's whats going on:
>
> Data inputs: ?9 different shapefiles (.shp) of different point locations
> (lat, long)
> ? ? ? ? ? ? ? ? ? shapefile bounding box ?(lat/long corner points (14, 1)
> (15,1) (14, 2) (15,2))
> ? ? ? ? ? ? ? ? ? 1 .csv of lat/long coordinates for points
>
>
>
> Goal: ?a- I want to randomly drop x number of points in a study area within
> the bounding box
> ? ? ? ? b- Then I want to add the .csv points to the mix
> ? ? ? ? c- Measure the distances from all these points to each of the 9
> shapefiles locations, one at a time and save the results in a table or list
> ? ? ? ? d- run an regression analysis on the table or list created (I have
> this code all set to go)
> ? ? ? ? e- I want to save the outputs of the model in a new table ?(I have
> this code also)
> ? ? ? ? f- ?Run this whole process again x number of times and save x
> number of outputs in the table in order of creation
>
> Any help on steps a, b,c, or f would be appreciated
>
> M.
>
>
> --
> View this message in context:
http://r.789695.n4.nabble.com/Randomized-Points-in-space-saving-model-results-tp3922612p3922612.html
> Sent from the R help mailing list archive at Nabble.com.
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>

On 21/10/11 06:32, magono wrote:
> A bit new to R and I'm working on something a bit more challenging than
I am
> used to- so here's whats going on:
>
> Data inputs:  9 different shapefiles (.shp) of different point locations
> (lat, long)
>                     shapefile bounding box  (lat/long corner points (14, 1)
> (15,1) (14, 2) (15,2))
>                     1 .csv of lat/long coordinates for points
>
>
>
> Goal:  a- I want to randomly drop x number of points in a study area within
> the bounding box
>           b- Then I want to add the .csv points to the mix
>           c- Measure the distances from all these points to each of the 9
> shapefiles locations, one at a time and save the results in a table or list
>           d- run an regression analysis on the table or list created (I
have
> this code all set to go)
>           e- I want to save the outputs of the model in a new table  (I
have
> this code also)
>           f-  Run this whole process again x number of times and save x
> number of outputs in the table in order of creation
>
> Any help on steps a, b,c, or f would be appreciated.
I  believe that the tools available in the "spatstat" package would be
of
use to you.  Read the vignette "Handling shapefiles in the spatstat 
package"
to find out how to convert shapefile objects to objects of classes with 
which
"spatstat" can deal.

     cheers,

         Rolf Turner

Thank you both for your help.  If I have any further questions after I give
this a go I'll get back to you.  This was much appreciated.

M.

--
View this message in context:
http://r.789695.n4.nabble.com/Randomized-Points-in-space-saving-model-results-tp3922612p3954834.html
Sent from the R help mailing list archive at Nabble.com.

Ok, to michael
Looking over what you posted I'm thinking this is the way forward but my
lack of experience using functions is getting in the way of accomplishing
the task.  Examining the outputs into the functions and putting in my own
info is giving me a headache.  Here's my thoughts at each step  marked with
``````-


runif2d <- function(n = 100, xmin = 1, xmax = 2, ymin = 14, ymax = 15){ 
```````put in my bounding box coordinates above
     stopifnot(all(xmax > xmin, ymax > min)) 
     y <- runif(n, ymin, ymax) 
     x <- runif(n, xmin, xmax) 
````````shall I enter the bounding box info again above in the y= and x= ? 
     cbind(x,y) }
`````Say I want to examine the resulting points created, how can I go about
this?

getdist <- function(inpoints, refpoints) { 
``````` which of these is my .csv of predetermined coordinates to add to the
mix and which of these points to the generated points above? 
    # Takes in two 2 column matrices representing (x,y) coordinates 
    #   and returns a matrix with all the distance pairs between them 
       stopifnot(all(is.matrix(inpoints), is.matrix(refpoints), 
dim(inpoints)[2L] == 2L, dim(refpoints)[2L] == 2L)) 
       d <- matrix(nrow = nrow(inpoints), ncol = nrow(refpoints)) 
       for (i in seq_along(refpoints)) { 
             d[, i] <- rowSums(inpoints-refpoints[i,]^2) 
       } 
       d 
} 
`````Again, I want to see the matrix created above with all the points and
distances, how do I call that up (I know this must be simple, but I think
since I can't get the whole shabang to work, I'm having trouble finding
the
right command to look at the results.)


Thanks so much by the way, you have an enviable command of R.

-M.


--
View this message in context:
http://r.789695.n4.nabble.com/Randomized-Points-in-space-saving-model-results-tp3922612p3954873.html
Sent from the R help mailing list archive at Nabble.com.