SSN (version 1.1.4)

BlockPredict: Block Predicton for Streams Data

Description

Block prediction for objects of class glmssn-class

Usage

BlockPredict(object, predpointsID)

Arguments

object

an object of class glmssn

predpointsID

a valid prediction points ID

Value

• A data.frame with one row and two columns. The first column, BlockPredEst, is the average prediction value, and the second column, BlockPredSE, is the standard error of the block prediction.

Details

This function operates on glmssn objects in much the same way as the predict function. BlockPredict uses the locations in the predpointsID data set to compute the average prediction value in the area defined by the prediction locations. These prediction locations are used to approximate the integral over that area, so they should be evenly spaced and dense in the area where block prediction is desired. The user needs to create these prediction locations and include them in the SSN object prior to fitting the model with glmssn.

References

Ver Hoef, J. M.. Peterson, E. E. and Theobald, D. (2006) Spatial statistical models that use flow and stream distance. Environmental and Ecological Statistics 13, 449-464. DOI: 10.1007/s10651-006-0022-8.

Examples

library(SSN)
# NOT RUN 
# mf04 <- importSSN(system.file("lsndata/MiddleFork04.ssn", 
#	        package = "SSN"), o.write = TRUE)
# use SpatialStreamNetwork object mf04 that was already created
data(mf04p)
#make sure mf04p has the correct path, will vary for each users installation
mf04p <- updatePath(mf04p, system.file("lsndata/MiddleFork04.ssn", package = "SSN"))

## NOT RUN Distance Matrix has already been created
## createDistMat(mf04)
# mf04p <- importPredpts(mf04p, "Knapp", "ssn")
# mf04p <- importPredpts(mf04p, "CapeHorn", "ssn")
names(mf04p)
## NOTE: need the amongpreds distance matrices for block prediction
#createDistMat(mf04p, predpts = "Knapp", o.write = TRUE, amongpreds = TRUE)
# just do CapeHorn Example
createDistMat(mf04p, predpts = "CapeHorn", o.write = TRUE, amongpreds = TRUE)

# NOT RUN see densely gridded prediction points on stream
# plot(mf04p, PredPointsID = "Knapp")

# NOT RUN   fit the model first
#fitSpBk <- glmssn(Summer_mn ~ ELEV_DEM + netID,
#    ssn.object = mf04p, EstMeth = "REML", family = "Gaussian",
#    CorModels = c("Exponential.tailup","Exponential.taildown",
#    "Exponential.Euclid"), addfunccol = "afvArea")
data(modelFits)
fitSpBk$ssn.object <- updatePath(fitSpBk$ssn.object, 
	system.file("lsndata/MiddleFork04.ssn", package = "SSN"))
# one-at-a-time predictions for CapeHorn stream
fitSpPredC <- predict(fitSpBk, "CapeHorn")
# NOT RUN plot densely gridded prediction points on stream
# plot(glmssn.BPCapeHorn, "Summer_mn")
# block prediction for CapeHorn stream
BlockPredict(fitSpBk, "CapeHorn")
## NOT RUN Another example
# one-at-a-time predictions for Knapp stream
#fitSpPredK <- predict(fitSpBk, "Knapp")
# NOT RUN plot densely gridded prediction points on stream
# plot(fitSpPredK, "Summer_mn")
# block prediction for Knapp stream
#BlockPredict(fitSpBk, "Knapp")