Great FAJ Article on Statistical Measure of Financial Turbulence

April 21, 2011

(This article was first published on Timely Portfolio, and kindly contributed to R-bloggers)

faj abstract

I particularly liked this well-written paper, since unlike most academic research, I was able to understand it, replicate it, and incorporate it.  I know that the Financial Analyst Journal is not considered by the academic community as a top-tier journal, but its insights have deepened my understanding and stimulated a lot of very beneficial thoughts.

As usual, I have struggled with how I could write this post with public data, but have not found an adequate public replacement for the CRB Index, so I will show how I used R to get similar results and build a basic system around it without including a file with the CRB monthly closes.  Everything else is publicly available, and I hope that even though the data is not included, the method and result will be very clear.

From TimelyPortfolio

When building the system around it, I was hesitant to do an optimization window with a walkforward as the article suggests, so with hindsight unfortunately, I picked a value (1) that works well.  In the next iteration, I achieve much better results with far less hindsight optimization bias by using correlation instead of covariance.  I’ll save that for my next post after a short weekend trip to the beach.

From TimelyPortfolio

R code:


#get data from St. Louis Federal Reserve (FRED)
getSymbols(“GS20″,src=”FRED”) #load 20yTreasury; 20y has gap 86-93; 30y has gap in early 2000s
getSymbols(“GS30″,src=”FRED”) #load 30yTreasury to fill 20y gap 86-93
#getSymbols(“BAA”,src=”FRED”) #load BAA
getSymbols(“SP500″,src=”FRED”) #load SP500

#get CRB data from a csv file

#fill 20y gap from discontinued 20y Treasuries with 30y

#do a little manipulation to get the data lined up on monthly basis
#get monthly format to yyyy-mm-dd with the first day of the month
#my CRB data is end of month; could change but more fun to do in R

#let’s merge all this into one xts object; CRB starts latest in 1956
#use ROC for SP500 and CRB and momentum for yield data
#get Covariances and multiply to by 100000 for 20y to sp500 and crb and 1000 for sp500 to crb to standardize
#don’t like this manual intervention; next post will use correlation instead
chartSeries(turbulence,theme=”white”,name=”Covariance and % Change as Measure of Financial Turbulence”)

#wish I could remember where I got some of this code
#most likely candidate is
#please let me know if you know the source
#so I can give adequate credit

#use turbulence to determine in or out of equal-weighted sp500 and crb
# Replace missing signals with no position
# (generally just at beginning of series)
signal[] <- 0

#get returns from equal-weighted crb and sp500 position; Return.portfolio was causing problems, so did the hard way
ret[1] <- 0

#get system performance
system_eq<- cumprod(1+ret*signal)


charts.PerformanceSummary(perf_comparison,ylog=TRUE,main=”Turbulence-based System vs Equal-Weighted CRB and SP500″)

To leave a comment for the author, please follow the link and comment on their blog: Timely Portfolio. offers daily e-mail updates about R news and tutorials on topics such as: Data science, Big Data, R jobs, visualization (ggplot2, Boxplots, maps, animation), programming (RStudio, Sweave, LaTeX, SQL, Eclipse, git, hadoop, Web Scraping) statistics (regression, PCA, time series, trading) and more...

If you got this far, why not subscribe for updates from the site? Choose your flavor: e-mail, twitter, RSS, or facebook...

Tags: , ,

Comments are closed.

Search R-bloggers


Never miss an update!
Subscribe to R-bloggers to receive
e-mails with the latest R posts.
(You will not see this message again.)

Click here to close (This popup will not appear again)