John Authers has been blogging from a CFA conference in Vancouver this week and had this post the other day about Nassim Nicolas Taleb's presentation that I was slow to find.Taleb had a couple of interesting thoughts about diversification. As quoted from the Alphaville post, "Diversification doesn’t work. You might need as many as 12,000 companies before it truly works."
Up front I should say that I view this differently than he does in this comment.
I take his comment to mean that he thinks owning the broadest of index funds is how to diversify and so buying the world makes the most sense, in this context. This is the crux of the passive indexer's argument of how to invest.
Of course this assumes the investor is interested in a diversified portfolio. As I read these comments about diversification I think about the first time I saw him (which was on the Connie Mack show) and he said to put 85-90% of your money into t-bills from around the world and go berserk (my word not his) with the remainder which implies taking a very undiversified approach, generally risk adverse but not a diversified equity portfolio.
Based on what I know of him (which may not be much) I would think he would prefer the mostly t-bill with a little bit of en fuego approach which I find to be far more interesting to ponder. If you had 90% of your money in t-bills, what would you put the 10% into?
The other nugget I found particularly interesting even if not new was the general idea, I am paraphrasing here, that volatility is manageable the vast majority of the time, more than the "than the predicted two-thirds of the time." But when volatility is not manageable is when people come unglued.
I don't necessarily think of volatility in the same way as he expressed it but it obviously rings true in that most of the time market chugs along smoothly (it has an up year close to 3/4 of the time) but people get very upset when the big declines come and obviously most folks don't see it coming.
It is this sort of thing, seeking out warnings like the 200 DMA (or similar) or the inverted yield curve, that becomes very important, as I see it anyway. They provide warnings of potential problems. This is a big focus of my practice and my writing. They will never let you quantify what might be coming but chances are you don't need quantify, you'd be happy enough to avoid it, or some of it anyway. If you know people freak out when the market rolls over into bear phase doesn't it then makes sense to watch out for when a bear phase might be starting?
Some obviously say no but not me.
23 comments:
Roger - market is set to trade above the 200 day SMA and EMA today. Are you starting to get bullish?
not in the least am i getting bullish.
however I will heed what the market says in baby steps.
what i usually do, and will do this time, is if it closes above the 200 dma today is see how the market is doing the next day and increase exposure, late in the day Monday if it looks like it will stay above on that next day.
I may it get it wrong but won't get whipsawed if i move slowly.
I agree. The rally this week is a little suspect....low volume, options expiration....we'll see though.
great blog by the way. you do an excellent job helping us do it yourself investors.
For the past few months I have been on several interviews in Boston and NYC whoring my wares as a Portfolio Manager with an "okay" book of business and a very successful track record. What has me concerned is every shop I went to (I am Value biased with a large contrarian bend but met with shops with all sorts of strategies) has been talking about playing the same themes - weak dollar, commodity prices. When I see a trend like this it I ususally go the opposite way. It's easy to post as an anon and call for the opposite of what everyone is reading, but thought I would share what I'm seeing out there and what has me tightening up international, and dare I say, going long US domestic? Again, it's easy to post from behind anon, but wanted to share.
di·ver·si·fi·ca·tion
2. the act or practice of manufacturing a variety of products, investing in a variety of securities, selling a variety of merchandise, etc., so that a failure in or an economic slump affecting one of them will not be disastrous.
The interesting conundrum for the investor is that diversification is not a static beast. Stocks (SPY) and commodities (DJP) have been moving in relative lock-step since early February. Were they both to sell-off together, a not-so-unlikely scenario, it would be beneficial for investors to game where the money could go. In March it was Treasuries (IEF hit ~$92.50 on March 20th), so that is one possibility. Are there others?
Roy that you say not static which answers the question. the relationship between stocks and commodities or any combo is that the correlation ebbs and flows. Two things have a typical correlation that stands up most of the time but occasional they deviate from normal.
Like when everything goes down during a panic, remember all of that sentiment last summer and fall.
Not perfect all the time but I think useful most of the time?
Or let it ride on red?
Roger, your commentaries had me spend more time looking for diversification in the order of using uncorrelated assets. I have bought RYFOX Rydex Alternative Strategies Allocation Fund which uses 4 of their Alternative assets funds as well as 1 Powershares currency Fund. Rather than make the decison on which funds, I'll leave that to Rydex. We'll see.
I read a lot on the internet about investing, some items are quite obscure. I always seem to focus my attention when a pattern is discovered, and there was one documented overlaying the recent S&P with an S&P chart (rebased) from George Bush's rein in the early 90s - I forget which site this is from - and the similarity was uncanny. The author also commented on there being a weak dollar, high oil price and uncertain credit markets back then too. From this chart the S&P is going to fall back to 1350 and then shoot for the moon. Well, around 1500 or so anyway. I'd be interested to see which sectors brought about this remarkable recovery and what happened to spark it? If oil fell to $60-80 and the credit markets returned to normal would retailers and even the banks suddenly be in vogue?
Regarding correlation, it seems Gold and Oil have diverged quite significantly from the norm. I know the primary reason is because I'm more heavily invested in the shiny stuff, but would this signal a coming reversal in the energy markets?
"Let it ride on red" - lol
Yes, I (my account, not the kids!) am currently letting it all ride on red. Obviously, the wheel has not been my friend this month.
I'm totally overwhelmed with this market...retired and insecure.
I read the following article.. how does a person decide which of these funds is right for them?
Are these to be considered core funds?
100% cash and thinking end of summer might be a good time to
invest long term.
thank you for your blog:-)
oops...here is the link to the article
http://biz.yahoo.com/ms/080508/237408.html?.v=1
to overwhelmed, don't be. maybe this will not resonate but there is nothing unique about the market's behavior through this. As of now the SPX is down 9.2% from its Oct high. That may not fit your idea of down a little but it doesn't seem like it is down a lot at this point. That might change or it might not but there is nothing new about down 10%. there would be nothing new about down 30% either.
as far as five funds for 15 years? ludicrous concept.
the like several funds for the manager. i am in no way taking a shot at any of the managers but a lot can happen to a manager in five years, Foster Freiss, Alberto Vilar, Bill Miller and the guy who liked Senetek, symbol STNKY, like stinky, was his name Heiko Thiemen? whatever five years is a long time in this context and they are going out 15 years, total disservice.
I have read all of Taleb's work and even dug up some stuff on his company and their investment strategies. I even named my own company "Black Swan Financial, Inc." because I liked the idea of the Black Swan and the fact that it symbolized small/unexpected things having large impacts (I am a one person shop).
Taleb used to run a fund that invested in very deep out-of-the-money options. Scooping up thousands of them for nickels and dimes with the expectation that the Black Swan events would be more frequent than the normal distribution implies and therefore the options would end up in-the-money more often than they were priced for. Not sure what happened to that (I stopped following him 18 months or so ago).
I went so far as to test his t-bills theory in a paper portfolio. I created a portfolio that was 90% SHY and then I picked 10 very high beta/high volatility stocks at random (1% each) from a list I have. Since the market top last October that portfolio has returned about 1.4% with a daily standard deviation about 1/5th that of the S&P 500. Granted this is an awfully short time period to judge but there are the results nonetheless.
While I find many of Taleb's ideas impractical from a portfolio management standpoint, I do have great respect for some of the ideas and incorporate them into my own risk assessments. I think drawing on off the wall ideas from him and people like Benoit Mandlebrot are an excellent sanity check for common market wisdom. It never hurts to apply game theory, chaos theory, prospect theory, and lots of other scientific endeavors to portfolio management...IMHO.
thanks for the advice Roger:-)
from overwhelmed
I also read both of Taleb's books, and his thesis was that the normal distribution curve was bogus. This negates the use of all financial ratios that are based on the statistical distribution of returns that financial planners use for asset allocation and returns forcasting. G. Considine did a good analysis on his website about why he disagrees with Taleb. As stated above, he was making bets that long shots would deliver his returns. You may not agree with his conclusions, but it's interesting to view the world through his eyes.
Another book worth reading is Arielly, Predictably Irrational.
Sam
My first time reading your blog.Found it on Forbes. Very informative.
Just found your blog after reading a post on www.summit-advisors.com After reading a few comments, I am going to re-read Taleb's book to look at it in a different light.
Rogers,
To avoid getting whipsawed around the 200 day MA, you could 1) set a moving average filter which averages data over time 2) or use a certain % above the 200 day MA on a closing basis filter to trigger your buy in.
An even better signal would have been the moving average crossover between the 50 day MA and the 200 day MA. (I noticed that that happened sometime end of December 2007)
What's your take on these thoughts?
Thanks,
-Sajal
of course it could work. whether this or something else should be done depends on what you are trying to achieve. I am trying to avoid a chunk of down a lot. The 200 DMA is very simple. What you describe adds a layer of complexity. for some this is appropriate and for some it is not. I am trying to keep this simple but you may have success with it.
roger,
thanks for this post; for me your timing seems uncanny. i just finished reading the black swan a few nights ago.
while taleb's book contains multitudes of fascinating ideas and wisdom, for investment purposes the essence seems to come down to his revealing a basic weakness in several investment gospels, and for me particularly, the black scholes option pricing model.
as he points out, black-scholes is mathematically elegant; however, in the real world the model fails. taleb's assertion to go 90% T-bills and 10% "berserk" reflects his desire to take advantage of the opportunity created by the deficiency in black-scholes. the problem with black scholes derives from its assumptions; black scholes rests on the assumption that stock returns tend to be "normally" distributed (in the book taleb refers to this as gaussian.) as long as returns behave according to a normal distribution, the black-scholes model tends to work well. taleb, however, realized that black swan (the unexpected) events force stock returns into non-gaussian patterns. when a black swan (such as october 19, 1987) occurs, the returns go 10 and 20 standard deviations from the mean. a five std deviation event should occur approx one every million observations; yet i looked at daily returns for the s&p 500 going back to the beginning of 1950 and i counted 24 days (out of 14685, or one out of every 583) where the return exceeded five std deviations. this alone seems to indicate that the returns do not fall into a normal or gaussian distribution. taleb indicates that the abnormal return of oct 19, 1987 should only occur once every several million years or so, but the world experienced two of those way-against-the-probabilty events during one century (and one century only includes around 22,000 or 23,000 trading days or events). the clear and real evidence indicates that stock returns do not follow a normal distribution. since black-scholes misprices underprices the value of options when a black swan occurs, taleb apparently made lots of small bets on options (the 10% "berserk" investment) to capture outsized returns relative to the option values. by making enough of these small (relative to the total value of one's portfolio) bets, taleb believes that one will hit enough black swans to capture better than market returns.
gjg49: Paul Wilmott has a set of thought provoking posts on the Black Scholes model (I have commented on these on my blog as well). Link to his posts is at:
http://www.wilmott.com/blogs/paul
I think that should make for an interesting counterpoint to Taleb's book. Didn't you feel he was maybe a little too pompous and arrogant?
Thanks,
-Sajal
Good Job! :)
Post a Comment