Keep rank over time in R - r

Question

Is there a way in R to check if ranks are kept over time for individual observations?
I have measured a number of plants over the years and want to check if large plants stay large and small plants stay small (i.e. if the large plants prevent other plants from growing). The plants are ranked 1-5 in size (from small to large).
I have measured about 1000 plants.
Very grateful for any answers or comments.
/Stina

Answer 1

Maybe you do something like this?
# create random data
plantId <- sample(1:50,1000,replace=TRUE)
rank <- sample(1:5,1000,replace=TRUE)
time <- as.POSIXct(sample(1000000:10000000,1000,replace=FALSE)[phone]*rank,origin="1970-01-01")
myData <- data.frame(plantId , rank, time )
# function to calculate the time a plant has a given rank
getRankTime <- function(id,testRank,data=myData){
plantData <- myData[myData$plantId==id,];
if(nrow(plantData) < 2){ # only one observed value of this plant
return(NA)
}else if(all(plantData$rank != testRank)){ # plant was never of the rank under consideration
return(NA)
}else{ # calculate the (censered) time the plant stay(ed) in rank 'testRank'
startObsTimeInRank <- min(plantData$time[plantData$rank == testRank])
if(any(plantData$rank > testRank)){
endObsTimeInRank <- min(plantData$time[plantData$rank > testRank])
}else{
#eighter take the last time
endObsTimeInRank <- max(plantData$time[plantData$rank == testRank])
# alternatively use the current time
# endObsTimeInRank <- Sys.time()
}
return(as.numeric(endObsTimeInRank - startObsTimeInRank))
}
}
# calculate the average time plants stay in a rank
allPlantIds <- unique(myData$plantId)
stayInRankTime <- list()
for(runRank in 1:5){
stayInRankTime[[runRank]] <- sapply(allPlantIds, function(runPlatId) getRankTime(runPlatId,runRank) )
}
# average time plants stay in acertain rank'
avgRankTime <- lapply(stayInRankTime,function(x)mean(x, na.rm =TRUE))
avgRankTime