Russell 2000 Softail Fat Boy

If the Russell 2000 were a motorcycle, maybe it should be a Harley-Davidson Softail Fat Boy.  I have explored the exceptional case of the Russell 2000 in quite a few posts

• More Exploration of Crazy RUT
• Where are the Fat Tails?
• Crazy RUT

but I still am not sure I have done a good job of clearly and simply explaining some of the unique characteristics of the Russell 2000 over the last  25 years.  The Russell 2000 has been extremely difficult to beat because the upside has been so good (fat boy) and the downside has been the same or better (softtail).

From TimelyPortfolio

From TimelyPortfolio

From TimelyPortfolio

R code from GIST:

	#explore exceptional case of the Russell 2000
	require(quantmod)
	require(PerformanceAnalytics)
	require(xtsExtra)
	getSymbols("^RUT", from = "1900-01-01")
	getSymbols("^GSPC", from = "1900-01-01")
	#do initial exploration of distribution
	chart.QQPlot(na.omit(ROC(RUT[,4],type="discrete",n=250))["1987::"])
	chart.QQPlot(na.omit(ROC(GSPC[,4],type="discrete",n=250))["1987::"])
	#explore barplot.xts to do a chart of annual returns for both indexes
	#merge prices
	prices <- merge(GSPC[,4],RUT[,4])
	#use endpoints to get annual returns
	returns.annual <- as.xts(apply(
	ROC(prices[endpoints(prices,"years")],type="discrete",n=1),
	MARGIN = 2,
	FUN = na.fill, fill = 0),
	order.by = index(prices[endpoints(prices,"years")]))
	#name columns something a little more clear
	colnames(returns.annual) <- c("S&P 500","Russell 2000")
	#using barplot.xts create the plot
	#I made some subtle changes to barplot.xts to experiment so plot will be cosmetically different
	barplot.xts(returns.annual,
	stacked=FALSE,
	box="transparent", #get rid of box surrounding the plot
	ylim=c(-0.5,0.5),
	ylab=NA,
	border=c(brewer.pal(n=11,"BrBG")[c(4,9)]),
	col=c(brewer.pal(n=11,"BrBG")[c(4,9)])) #deliberately trying some new colors
	title(main="Annual Returns of S&P 500 and Russell 2000",
	outer = TRUE,
	adj=0.05, font.main = 1, cex.main = 1.25, line = -2)
	require(latticeExtra)
	require(reshape2)
	roc <- na.omit(merge(ROC(GSPC[,4],type="discrete",n=250),ROC(RUT[,4],type="discrete",n=250)))
	#name columns something a little more clear
	colnames(roc) <- c("S&P 500","Russell 2000")
	roc.melt <- melt(coredata(roc))
	asTheEconomist(
	densityplot(~value,data=roc.melt,groups=Var2,
	#par.settings=theEconomist.theme(box="transparent"),
	#lattice.options=theEconomist.opts(),
	auto.key=list(space="right",col=(brewer.pal(n=11,"BrBG")[c(9,4)]),lines=FALSE),
	col=(brewer.pal(n=11,"BrBG")[c(9,4)]),
	ylab=NA,
	main="Annual Returns (Rolling 250 Day) of S&P 500 and Russell 2000")
	)

view raw fat tails to the upside.r hosted with ❤ by GitHub