#This script interpolates station value for the Oregon case study. This program loads the station data from a csv file and perform two types  of regression: multiple linear model and general additive model (GAM). Note that this program 1)assumes that the csv file is in the current working 2)extract relevant variables from raster images before performing the regressions. The user must provide the list of raster images

#in  a textile.

#Created by Benoit Parmentier on February 12, 2012.

library(raster)                                                                              # loading the raster package

library(gtools)                                                                              l# loading ...

#list_var<-list.files()                                                                        # displaying the files in the current folder which should contain only the relevant raster images.

#dump("list_var", file="list_var.txt")                                                # Saving the values of list_var in a ascii text file.

ghcn<-read.csv(paste(path,"/",infile1, sep=""), header=TRUE)    #The "paste" function concatenates the path and file name in common string.

names(ghcn)                                                                                 #Checking that the columns are correctly labelled.

coords<- ghcn[,c('x_OR83M','y_OR83M')]                                    #Selecting all rows for the x and y columns and packaging the output in a data frame.

#s_raster<- stack(inlistvar)                                                              #Creating a stack of raster images from the list of variables.

#stat_val<- extract(s_raster, coords)                                               #Extracting values from the raster stack for every point location in coords data frame.

###Creating a validation dataset by creating training and testing datasets (%30)

n<-nrow(ghcn1507)

ns<-n-round(n*0.3)  #Create a sample from the data frame with 70% of the rows

ind.training <- sample(nrow(ghcn1507), size=ns, replace=FALSE) #This selects the index position for 70% of the rows taken randomly

ind.testing <- setdiff(1:nrow(ghcn1507), ind.training)

ghcn1507_s <- ghcn1507[ind.training, ]

ghcn1507_v <- ghcn1507[ind.testing, ]

#ghcn1507_s<- sample(ghcn1507,size=ns, replace=T)

ghcn1507_s<- ghcn1507[sample(nrow(ghcn1507), size=ns), ]

#ghcn1507_v <- ghcn1507$STAT_ID[ !ghcn1507_s$STAT_ID ] #dataset for validation that can be used to predict values.