# LIBRARIES
library(openxlsx) 

# INPUT THE PATH AND FILE NAME OF TARGET XLSX
FilePath <- '/Users/SQIANG/Desktop/'
FileName <- "RawData S. Qiang 180424"
InputFileType<-".xlsx"
FileID<-paste(FileName, InputFileType, sep = "")

# INPUT SHEET NUMBER (e.g. 1, 2 and 3)
SheetNum = 3

# READ DATA
RawData <- read.xlsx(paste(FilePath, FileID, sep = ""), sheet = SheetNum, colNames = F)

# DETERMINE THE NUMBER OF Columns OF RawData
SampleSize <- ncol(RawData)
DataSize <- nrow(RawData)-1
GroupEnd <- matrix(0,nrow=SampleSize,ncol=1)

# DEFINE GROUP #
for(i in 1:SampleSize){
	temp1 = RawData[1:DataSize+1,i]
	temp2 = which(is.na(temp1))
	GroupEnd[i] = temp2[1]
	if (is.na(temp2[1])) GroupEnd[i] = DataSize+1
}

# CALCULATION AND OUTPUT
for (i in 2:SampleSize){
	A_raw = RawData[2:GroupEnd[i],i]
	A_raw = as.numeric(as.character(A_raw))
	A=sort(A_raw, decreasing = T)
	S = GroupEnd[i]-1
	temp = c(1:S)
	x = c(1:S)
	y = c(1:S)
	At = sum(A)
	tempp1 = 0
	tempp2 = 0
	for (j in 1:S){
		qr_ = A[j]/A[1]
		x[j] = log(temp[j])
		y[j] = log(qr_)
		tempp1 = tempp1 + x[j]*y[j]
		tempp2 = tempp2 + x[j]*x[j]
	}
	p = -1*tempp1/tempp2
	sstotal = 0
	ssresid = 0
	my = mean(y)
	for (j in 1:S){
		sy = -1*p*x[j]
		sstotal = sstotal + (y[j]-my)**2
		ssresid = ssresid + (sy-y[j])**2	
	}
	ssreg = sstotal - ssresid
	R2 = ssreg/sstotal
	if (i == 2) {
		output = list (RawData[1,i], p, R2)
		}
	else {
	tempp = list(RawData[1,i], p, R2)
	output = rbind(output,tempp)
	}
}

# OUTPUT THE FILE NAME OF TARGET XLSX
OutputFileName = "OutFile - "
OutputFileName = paste(OutputFileName, FileName, sep="")
OutputFileType = ".csv"
OutputFileID = paste(OutputFileName, OutputFileType, sep="")

# OUTPUT THE FILE INCLUDING "Sample Name", "p" AND "R2"
colnames(output)<-c("Sample Name", "p", "R2")

# WRITE INTO THE FILE
write.csv(output, paste(FilePath, OutputFileID, sep=""), row.names=F)