[R-sig-Geo] random spatial distribution code / shapefiles with more than one polygon

[Abby Rudolph]

Hi,

I was given the following code to determine whether the spatial distribution of attribute=1 is significantly different from the spatial distribution of attribute=0
given that the points (overall) are not randomly distributed in space.


BaltCity2<-readShapePoly("baltcity_mdsp83m.shp")
summary(BaltCity2)

x<-BaltCity2 at polygons[[1]]@Polygons 
x 
summary(x)
# this shows that the baltimore polygon is a single shape
# the Baltimore shapefile is projected in meters, so the following code converts meters to miles
coords<-slot(x[[1]],"coords")
coords<-coords[167:1,]
coords<-coords[-1,]
coords.mi<-coords*0.000621371192 
summary(coords.mi)

play2<-ppp(play$X_m*0.000621371192,play$Y_m*0.000621371192,
           window=owin(poly=list(x=coords.mi[,1],y=coords.mi[,2])))

I am able to get the program to work when I use the Baltimore shapefile because the polygon is a single shape.  

However, when I apply this code to data I have collected from New York City, I have problems.  I am wondering if the problem is the result of the NYC shapefile having multiple polygons and water gaps.  I have been attempting this code with two different NYC shapefiles.  The first is from NYC Bytes.  I brought it into ArcMap and projected it into meters. 
When I plot the following in R, the points from my data file map onto the shapefile and both are in meters

plot(NYCBB)
u<-RDS_TSO$RDS_new==0
points(RDS_TSOm$X_m[u],RDS_TSOm$Y_m[u],pch=5,col="yellow")
points(RDS_TSOm$X_m[!u],RDS_TSOm$Y_m[!u],pch=5,col="red")

However, when I use the code below to convert the meters to miles:

x<-NYBorough at polygons[[1]]@Polygons
x
# there are 8910 rows in x
summary(x)
# x is 5 different shapes compared to one like in the baltimore example above
     Length Class   Mode
[1,] 1      Polygon S4  
[2,] 1      Polygon S4  
[3,] 1      Polygon S4  
[4,] 1      Polygon S4  
[5,] 1      Polygon S4  

coords<-slot(x[[1]],"coords")
coords
# Now there are only 47 rows - with the baltimore data there was no reduction - I suspect it is being limited in some way
coords<-coords[19:1,]
coords
coords<-coords[-1,]
coords
NYBorough.coords.mi<-coords*0.000621371192 
NYBorough.coords.mi

then, when I try to convert my data x,y coordinates from meters to miles and create a shape with the NYBorough owin, I get a warning that all of the points fall outside the window

RDS<-ppp(RDS_TSOm$X_m[u]*0.000621371192,RDS_TSOm$Y_m[u]*0.000621371192,
           window=owin(poly=list(x=NYBorough.coords.mi[,1],y=NYBorough.coords.mi[,2])))

I received the following warning:
Warning message:
In ppp(RDS_TSOm$X_m[u] * 0.000621371192, RDS_TSOm$Y_m[u] * 0.000621371192,  :
  217 points were rejected as lying outside the specified window

This is puzzling, because when they are plotted using the following code, they overlap

plot(NYBorough)
points(RDS_TSOm$X_m[u],RDS_TSOm$Y_m[u],pch=5,col="yellow")

and there are only 217 points in this group, so it appears that they are all rejected.

I've also tried downloading the TIGER files from the US Census and merging all 5 boroughs, projecting the data in arcmap into meters and opening in R.
I get the same warning with both shapefiles.

Does it seem like the error is resulting from the shapefile I am using or some other error in the code?

Any help would be very much appreciated.

Thanks!

Abby