Oct
17
Correlation Update, from Phillip J. McDonnell
October 17, 2008 |
The volatility of the market has dramatically increased in the last few weeks. It is often said that the correlations between various markets increase at such times due to forced margin selling. Certainly such a suggestion is plausible on its face but like everything else must be tested.
One can look at the same-day correlations between various macro variables and the S&P. For this purpose the relevant ETFs were chosen. The correlations with SPY are as follows:
Oil 81%
Gold -32
Tbonds -53
Tbill -53
Yen -64
UK stocks 93
Japan stocks 93
VIX -86
The most striking is the strong positive correlation with oil. One interpretation is that oil is driven by recession fears just as stocks are. Another explanation may be that oil is being liquidated to finance stock margin accounts just as the pundits claim. Clearly holding oil is not now a hedge against a stock portfolio.
But when we look at gold the correlation is negative. This would tend to serve as evidence against a wholesale correlation of assets being sold. To some extent gold is still a hedge against a stock portfolio. The same goes for treasury paper and the yen. But we do see UK and Japanese stocks being strongly positively correlated. So it appears that many world markets are strongly correlated with each other. Again this may be a sign of coordinated margin liquidation. The strong negative VIX correlation can be interpreted as the markets are now being strongly driven by fear. None of this is predictive but is an interesting descriptive look at where we are now.
Dr. McDonnell is the author of Optimal Portfolio Modeling, Wiley, 2008
Comments
3 Comments so far
Archives
- May 2013
- April 2013
- March 2013
- February 2013
- January 2013
- December 2012
- November 2012
- October 2012
- September 2012
- August 2012
- July 2012
- June 2012
- May 2012
- April 2012
- March 2012
- February 2012
- January 2012
- December 2011
- November 2011
- October 2011
- September 2011
- August 2011
- July 2011
- June 2011
- May 2011
- April 2011
- March 2011
- February 2011
- January 2011
- December 2010
- November 2010
- October 2010
- September 2010
- August 2010
- July 2010
- June 2010
- May 2010
- April 2010
- March 2010
- February 2010
- January 2010
- December 2009
- November 2009
- October 2009
- September 2009
- August 2009
- July 2009
- June 2009
- May 2009
- April 2009
- March 2009
- February 2009
- January 2009
- December 2008
- November 2008
- October 2008
- September 2008
- August 2008
- July 2008
- June 2008
- May 2008
- April 2008
- March 2008
- February 2008
- January 2008
- December 2007
- November 2007
- October 2007
- September 2007
- August 2007
- July 2007
- June 2007
- May 2007
- April 2007
- March 2007
- February 2007
- January 2007
- December 2006
- November 2006
- October 2006
- September 2006
- August 2006
- Older Archives
Resources & Links
- The Letters Prize
- Pre-2007 Victor Niederhoffer Posts
- Vic’s NYC Junto
- Reading List
- Programming in 60 Seconds
- The Objectivist Center
- Foundation for Economic Education
- Tigerchess
- Dick Sears' G.T. Index
- Pre-2007 Daily Speculations
- Laurel & Vics' Worldly Investor Articles
The dangers of using historic correlation during a deleveraging environment can be seen in a global asset allocation portfolio. I invest a portion of retirement assets in a twenty-four component portfolio that is internally weighted for an expected annual 10% return at 8% standard deviation. Recently many of the portfolio components have become highly correlated, most strikingly during the two-weeks ending Oct. 14. At one point, the sum of intra-portfolio correlations, using 60-days data, spiked to 42% with a 31% standard deviation. However, even in short restricted account there is a solution. With the advent of short ETFs, simply establish an effective short position, (SDS for this example, which has a negative 51.5% intra-portfolio correlation) at a level calculated to bring the sum of intra-portfolio correlations into the desired range (in this case + or – 6%). This short position brings expected returns and volatility back in line with model expectations. Be prepared to adjust frequently though, as relationships are beginning to return to a more normal
anything normal about this trend.
3 consecutive trade days data for 3 years……
date….#of Companies….Total Vol…Block Vol…Totl Trades
block trades
2008
19-Sep 3200 4,065,219,165 496,162,633 13,022,420 14,344
18-Sep 3199 3,990,786,026 438,886,083 15,563,948 13,012
17-Sep 3194 3,240,667,427 443,570,648 12,782,761 10,874
2007
19-Sep 3,281 2,234,189,510 329,894,832 7,175,239 10,734
18-Sep 3,281 2,157,640,265 314,998,802 6,981,634 10,109
17-Sep 3,284 1,561,091,236 314,127,024 4,732,082 7,799
2006
20-Sep 3344 2,253,390,918 349,210,437 5,573,322 15,464
19-Sep 3334 2,160,872,017 389,412,160 5,748,431 13,849
18-Sep 3336 1,980,975,019 464,671,481 4,851,228 11,448
Can you please tell me how a speculator uses this information. It seems like a bunch of fluff, but I don’t have a PhD…