问题
I am trying to model a point process with an image covariate using the ppm() function in the spatstat package in R. I convert my raster to an im object for use with spatstat, and I run into a problem using the im as a covariate in the model. The pixel values are numeric, but these are actually just codes for different landscape zones so the crux of the problem is getting the model to read the pixel values as factor rather than numeric. I have tried the following two approaches (R code and data are presented below). The first consists of converting the raster values from numeric to factor prior to converting the raster object to the im object. Using the as.factor() function this seems to have the desired effect of converting the values to factor. However, when I run the ppm model with this covariate, the ppm() function does not include a parameter for each factor level in the model (compared to a reference level). Rather it treats the covariate as numeric with just the one parameter for the one covariate. The second approach was to run the ppm model with factor(covariate) used to specify the covariate in the formula argument, rather than just the covariate itself. This actually works in fitting the model, giving me a parameter for each factor level compared to the reference. However, when I run predict.ppm() to get my predictions it fails because I used factor() in the formula argument. The qustion is, how can I run the ppm model such that it recognizes the values of the covariate image as factor and, thus, fitting a model with a parameter estimate for each factor level (minus the reference) and allowing prediction with predict.ppm().
The point process data is in csv format here: https://www.dropbox.com/s/tp1opzsmc14e2hb/EbolaData_AnalyticSet_8.8.14.csv?dl=0
The tiff file for the covariate is here: https://www.dropbox.com/s/0fyt0jflokrpp5z/anthrome2000_global_5min.tif?dl=0
And the R code is as follows:
library(raster)
library(spatstat)
library(geostatsp)
# First set the geographic extent we'll be using
e <- extent(-20, 60, -40, 35)
# Then read in the point process data in .csv file:
outbreaks <- read.csv("EbolaData_AnalyticSet_8.8.14.csv")
coordinates(outbreaks) <- ~Longitude+Latitude
proj4string(outbreaks) <- CRS("+proj=longlat +ellps=WGS84 +datum=WGS84 +no_defs")
# Then read in the anthropogenic biome tiff and crop it
anthro2000 <- raster("anthrome2000_global_5min.tif")
anthro2000.africa <- crop(anthro2000, e)
# Then we define oour point process and window for spatstat:
SP <- as(outbreaks, "SpatialPoints")
outbreaks.ppp <- as(SP, "ppp")
# Now let's create a window
SP.win <- as(e, "SpatialPolygons")
W <- as(SP.win, "owin")
# Before creating the im object, let's convert the pixel values in raster to factor:
is.factor(anthro2000.africa)
f <- as.factor(anthro2000.africa)
is.factor(f)
rat <- levels(f)[[1]]
rat$anthro <- c("Urban", "Mixed Settle", "Rice Villages", "Irrigated villages", "Rainfed villages", "Pastoral vilalges",
"Resid. irrig. cropland", "Resid. rainfed cropland", "Pop. cropland", "Remote cropland",
"Resid. rangeland", "Pop. rangeland", "Remote rangeland", "Resid. forests", "Pop. forests",
"Remote forests", "Inhabited treeless and barren", "Wild forests", "Wild treeless and Barren")
rat$code <- c(11,12,21,22,23,24,31,32,33,34,41,42,43,51,52,53,54,61,62)
levels(f) <- rat
# Now let's convert that raster to an im object for use in spatstat:
anthro2000.africa.img <- asImRaster(f)
# Everything is now set up for the ppm models
# Aprroach 1
spatial.m.1 <- ppm(outbreaks.ppp, ~ Cov1, covariates = list(Cov1 = anthro2000.africa.img))
spatial.m.1 # Notice the model is fitted, however the pixel values of the covariate are not interepreted as factor
# Approach 2:
spatial.m.2 <- ppm(outbreaks.ppp, ~ factor(Cov1), covariates = list(Cov1 = anthro2000.africa.img)) # Noticce the use of factor() here to force the covariate as factor
spatial.m.2 # Here the model is fitted correctly with covariate as factor and thus each factor level has a parameter estimate in the model (relative to the ref)
# However, the approach does not allow me to run the predictions:
p <- predict.ppm(spatial.m.2, covariates = list(Cov1 = anthro2000.africa.img))
回答1:
The problem is that R doesn't have factor valued matrices, so it always a bit quirky to get factors into an im, but once it is done everything works as it should. My solution was just to convert the integer valued raster into the im format and handle everything from there (I'm not a regular user of the raster package).
I had to load the maptools library for the command SP <- as(outbreaks, "SpatialPoints") to work. Also, R couldn't read the given csv file due to some strange characters in the first column (which we don't use anyway), so I had to remove these for everything to work.
The syntax below for ppm requires that you are running spatstat 1.37-0 or newer. Furthermore, I'm using a new generic function Window from 1.38-0, and you need to do this slightly differently for older versions (I highly recommend the newest version 1.38-1):
library(raster)
library(spatstat)
library(geostatsp)
library(maptools)
# First set the geographic extent we'll be using
e <- extent(-20, 60, -40, 35)
# Then read in the point process data in .csv file:
outbreaks <- read.csv("EbolaData_AnalyticSet_8.8.14.csv")
coordinates(outbreaks) <- ~Longitude+Latitude
proj4string(outbreaks) <-
CRS("+proj=longlat +ellps=WGS84 +datum=WGS84 +no_defs")
# Then we define our point process (with the bounding box as temporary window)
# for spatstat:
SP <- as(outbreaks, "SpatialPoints")
outbreaks.ppp <- as(SP, "ppp")
# Then read in the anthropogenic biome tiff and crop it
anthro <- raster("anthrome2000_global_5min.tif")
anthro <- crop(anthro, e)
# Now let's convert that raster to an im object for use in spatstat
# (and make it into a factor):
anthro <- asImRaster(anthro)
anthro <- eval.im(as.factor(anthro))
levels(anthro) <-
c("Urban", "Mixed Settle", "Rice Villages", "Irrigated villages",
"Rainfed villages", "Pastoral vilalges", "Resid. irrig. cropland",
"Resid. rainfed cropland", "Pop. cropland", "Remote cropland",
"Resid. rangeland", "Pop. rangeland", "Remote rangeland",
"Resid. forests", "Pop. forests", "Remote forests",
"Inhabited treeless and barren", "Wild forests",
"Wild treeless and Barren")
# Make Africa into the observation window (of type mask):
Window(outbreaks.ppp) <- Window(anthro)
# See the data we have read in:
plot(anthro)
plot(outbreaks.ppp, add = TRUE)
# Fit model and predict:
spatial.m.1 <- ppm(outbreaks.ppp ~ anthro)
p <- predict.ppm(spatial.m.1)
来源：https://stackoverflow.com/questions/26162955/r-package-spatstat-how-to-use-point-process-model-covariate-as-factor-when-pixe