Supplementary_Functions.R

###---------SUPPLEMENTARY FUNCTIONS

library(reshape2)


###---------Count NAs
count_nas <- function(x) length(which(is.na(x)))

###---------Data Extraction
extract <- function(data, columns=NULL, fun = NULL, xvar, listvar){
  if (is.null(fun)) { #only raw data?
    a <- subset(data, select=columns)
    return(a)}
  else {
    indexlist <- list()
      for (i in 1:length(listvar)){ # tapply needs as a list
        indexlist[[i]] <- data[,listvar[i]]
      }
    b <- t(tapply(X=data[,xvar[1]], INDEX = indexlist, FUN=fun, na.rm = T))
    b <- melt(b)
      if (length(xvar) > 1){ # if several columns should be processed
        for (i in 2:length(xvar)){
          c <- t(tapply(X=data[,xvar[i]], INDEX = indexlist, FUN=fun, na.rm = T))
          c <- melt(c)
          b <- cbind(b, c[,NCOL(c)])
        }
      }
    names(b) <- names(data[,c(rev(listvar),xvar)])# for some reason, tapply always reverses the order of listvar    
    return(b)}    
}

#problems left: 
#1. names might not be assigned correctly
#2. if one of the xvars is shorter than the one processed before because ofna.rm, that cell is still assigned NA
#   also if there are rows missing due to NAs in first column -> values of other columns might be lost
#   SOLUTION: y(including NAs)~AllMeans$year[which(!is.na(AllMeans$y))]

# tested via: mean(mules1962$d3[which(mules1962$year == 1962 & mules1962$macrounit == "0-1")])

###-------- One plot per macrounit displaying gam-object

macrounitplots <- function(glmobject, ycol="MDperKMsqFall_mean", xcol="year",title, colour="red", shape="l"){
  par(mfrow=c(2,2),oma=c(2,0,2,0))
  macrounits <- levels(glmobject[,"macrounit"])
  for (i in 1:length(macrounits)){
    cond = which(glmobject[,"macrounit"]==macrounits[i])  
    plot(AllMeans[cond,ycol]~AllMeans[cond,xcol], type="p", main=macrounits[i], xlab=xcol, ylab=ycol)
    
    lines(x=AllMeans[cond,xcol], glmobject[cond,"fit"], col=colour, type=shape)

    #lines(x=AllMeans[cond,xcol], glmobject[cond,"fit"]  + 2 * glmobject[cond,"se.fit"], col=colour, lty=2, type=shape)
    #lines(x=AllMeans[cond,xcol], glmobject[cond,"fit"]  - 2 * glmobject[cond,"se.fit"], col=colour, lty=2, type=shape)
  }
  title(as.character(title), outer=TRUE)
}

# maybe add:
#yrange <- c(min(AllMeans[,ycol], na.rm=TRUE),max(AllMeans[,ycol],na.rm=TRUE)) 
#ylim=yrange
#for better comparability between macrounits(not always helpful)


### ------------------ functions for pairs()
panel.cor <- function(x, y, digits = 2, prefix = "", cex.cor, ...)
{
  usr <- par("usr"); on.exit(par(usr))
  par(usr = c(0, 1, 0, 1))
  r <- abs(cor(x, y, use="pairwise.complete.obs"))
  txt <- format(c(r, 0.123456789), digits = digits)[1]
  txt <- paste0(prefix, txt)
  if(missing(cex.cor)) cex.cor <- 0.8/strwidth(txt)
  text(0.5, 0.5, txt, cex = cex.cor * r)
}

panel.hist <- function(x, ...)
{
  usr <- par("usr"); on.exit(par(usr))
  par(usr = c(usr[1:2], 0, 1.5) )
  h <- hist(x, plot = FALSE)
  breaks <- h$breaks; nB <- length(breaks)
  y <- h$counts; y <- y/max(y)
  rect(breaks[-nB], 0, breaks[-1], y, col = "cyan", ...)
}