teaser/stock_return_bins/histStock.R

listIn <- commandArgs(TRUE)
library(lubridate)

data <- read.csv("/path/to/assignment/explainer")

myHist <- function(df, listIn)                 #listIn has metric[1-numerator & 2-denominator], minValue[3], and number of months for price return[4]
{
	nbins <- 100
	df2 <- subset(df, select= c("ticker_exchange", "price", "date"))
	df2$date <- as.Date(df2$date, format="%d/%m/%Y")
	df2$date <- format(df2$date, "%m-%Y")

	if (listIn[2]=='1')
	{
	df <- subset(df, select=c("ticker_exchange", "price", "date", listIn[1]))
	df <- df[complete.cases(df), ]
	df[, 4] <- as.numeric(as.character(df[, 4]))
	df <- subset(df, df[, 4] > as.numeric(listIn[3]))
	}
	else
	{
	df <- subset(df, select=c("ticker_exchange", "price", "date", listIn[1], listIn[2]))  # filtered unnecessary columns
	df <- df[complete.cases(df), ]
	df[, 4] <- as.numeric(as.character(df[, 4]))
	df[, 5] <- as.numeric(as.character(df[, 5]))
	df <- subset(df, (df[, 4]/df[, 5]) > as.numeric(listIn[3]))
	}
#============================= applied constraint for metric above ===================================
	df$date <- as.Date(df$date, format="%d/%m/%Y")                                             # converted to date object
	df$nextDate <- df$date %m+% months(as.integer(listIn[4]))                             # found next date for calculating returns
	df$date <- format(df$date, "%m-%Y")
	df$nextDate <- format(df$nextDate, "%m-%Y")                                           # changed format of both for correct m-Y matching
	df <- subset(df, select=c("ticker_exchange", "price", "nextDate"))

	outframe <- merge(df, df2, by.x=c("nextDate", "ticker_exchange"), by.y=c("date", "ticker_exchange"))
#============================== arranged 2 prices at 2 dates by merging along ticker and dates =========
	colnames(outframe)[3] <- "P2"
	colnames(outframe)[4] <- "P1"
	outframe[, 3] <- as.numeric(as.character(outframe[, 3]))
	outframe[, 4] <- as.numeric(as.character(outframe[, 4]))

	outframe$returns <- ((outframe$P1/outframe$P2) -1)*100                                   #calculated returns
	outframe <- unique(outframe[complete.cases(outframe), ])                                 #removed duplicate entries and NAs

	range <- max(outframe$returns)-min(outframe$returns)
	bins <- seq(min(outframe$returns)-(range/100), max(outframe$returns)+(range/100), by=range/nbins)
	hist(outframe$returns, breaks=bins)
	return(transform(table(cut(outframe$returns, bins))))
}

myHist(data, listIn)