I recently loaded up on 2x inverse funds yesterday aftenoon in fear and now I am missing out on the gains. I give up on this manipulated crap.
Wow. This brings up several points that are always worth revisiting. If you never take any defensive steps in your investment lifetime (30-40 years minimum) you save properly, and you allocate properly over time you have an outstanding chance of ending up with enough money.
The next point is that the stock market has an up year 72% of the time.
These are two very important big macro points that investors need to be cognizant of before devising exit strategies, trading philosophies or anything else. If you bought the S&P 500 the two days before the 1987 crash (the market was down about 2% the day before the crash) and just sat tight for ten years you would have more than tripled your money.
Sitting tight is not easy but this is the back drop.
The decline that the market is working through right now is well within the realm of down a little and normal stock market volatility. It is very unlikely that an investor can regularly game 5% moves in the market. The notion of loading up (although this is not defined in the comment) on inverse funds is a speculative one and with apologies to the reader gives a lesson to learn by.
The only context I have ever touched on with these is as a means of reducing exposure. As a hedge they are very simple and not disruptive to the portfolio.
Another fundamental belief is that I would not load up on anything. I still see commentary advising 20%-ish in gold which to me seems to introduce a ton of volatility to a portfolio except when priced in the Malawi kwacha (Peter Schiff heckle).
One thing I tell people every now and then about what I do with portfolios is that the market goes up most of the time, I'm just trying not get in the way and mess that up.
The defensive action I took last summer was poorly timed but not extreme and so the consequence was a lag for a few months. From the bottom in July to year end the SPX rallied about 15%. Lagging a move like that is not detrimental to the long term, missing it could be.
Most people do not need to manage against a one year time period let alone one week.
20 comments:
World Beta references a worthwhile read, I think
ttp://www.copstrat.com/COPWP_
Strat.html
Perhaps more up tomt's alley. I do not think, on cursory scan, that this is written by someone dogmatic to a rotation strategy. Just someone, like us, not in the biz, trying to improve returns, and a bit of fresh air. In my ideal world, the taxable account has a rogeresque diversifed portfolio while the tax advantaged account uses rotation.
Off topic question: Is there a way to subscribe to your columns on smartmoney.com/realmoney.com? Esp. via an RSS reader?
Thanks,
Stephen
the RSS feed should be
http://randomroger.blogspot.com/rss.xml
Heh, no, I'm subscribed to you blog. I meant this stuff (if not, I guess I've got a new website to check daily):
http://find.thestreet.com/cgi-bin/texis/rmauthor/?au=A1100633
oh i get it, no the blog feed is not integrated into TSCM/RM. From their end I doubt they want that, from my end the compliance is probably a night mare.
Makes sense - thanks.
Good post.
John Barber and Dan Laimon authored a piece for the Journal of Retirement Planning (Nov/Dec 2006) entitled: MANDATING THE PROBABILITY OF SUCCESS: A NEW APPROACH TO RETIREMENT PLANNING.
Without taking space here to summarize the document,I highly
recommend it. The study is the best roadmap towards retirement planning I have read, can be understood by just about anyone and is only ten pages in length.
Roger gave a link to this a few months ago. Another service provided by this excellent blog.
Thanks T, funny John and Dan are personal friends of mine.
for the paper T refers to click here
Anon, thanks for the Copstrat link.
I skimmed it and it looks interesting. His strategy is very basic and easy to execute - a good thing. It's a little riskier than what I do. It looks like he updates his holdings monthly and is always 100% in some combination of 10-15 asset classes. This is fine if you can stomach the drawdowns.
I'm amazed there aren't more people using momentum/rotation strategies - for at least a part of their portfolio. The evidence that this stuff works is overwhelming.
Well, if you know that the market is heading lower in the short-term, it makes sense to move into cash 100%.
My retirement funds are invested in Vanguard's S&P 500 index fund, so with one quick transaction I can move to 100% money market in 5 seconds. Once things stabalize, I can get back in.
I found a comment made by Geoff Considine (QUANTEX) quite surprising, to wit
"In general, this portfolio seems to have a lot of “moving parts,” i.e.
a number of individual allocations at very small percentages.
I do not see a lot of value in a 1%-2% allocation to something."
(http://biz.yahoo.com/seekingalpha/070307/28934_id.html?.v=1)
I did not think a 1-2% allocation to a particular holding to be of no value.
Where do you come down on this issue? I appreciate your comments.
Jay Charles
the number of times the market whooshes up for no reason at all exactly when it shouldn't makes 100% cash wildly aggressive.
Timing Model update:
Nothing much has changed except the 5 dma of Value Line Index is now a tad more below it's 75 dma. Tomorrow will be a pivotal day imo. I'm currently a fence sitter at 45% long. If we have a big up day, I'll be buying a little, a big down day, I'll be selling a little, or if we don't see much direction - I'll stay put, but will take that as a bearish omen.
Note, we are still far above the 250 dma. The long term term is decidedly up.
The sentiment models are still cycling down but far from signaling blood-in-the-street fear.
Anon 4:50,
If you're in an index fund, you don't move into cash (or back) right away. You get the end of the day close; now an ETF is different.
Also, many of Vanguard's funds - not the money markets mind you - mandate a two month waiting period before you can switch out (or back) penalty free.
I would try pretending to time before you really do. See what your results are at guessing the markets ups and downs.
It's not so easy.
…next point…
I'm a huge Considine - and MVO-type calculations - fan. Mean Variance optimization, at the very least, gets you thinking in terms of the portfolio, in terms of risk, and in terms of asset class correlation.
The statistics show that after twenty stocks you don't get a diversification benefit. So twenty stocks means a five percent minimum allocation in an all stock portfolio, much much more than the one or two percent he eschews.
He also says three individual stocks are not enough; so you have to find four to nineteen stocks!
Oversimplifying, but, a five percent drop in your non-individual stock portfolio - if it made up 97% of the total - would mean that if the other 3% doubled in you'd still be negative. So with such a small allocation, you'd need that “little allocation” going in the other direction in a big way to have any diversification effect.
The problem with this added individual stock picking is obvious: it's hard to find stocks to use. Just because they have an appropriate beta and R^2 doesn't really help when management back dates their options or develops the next Betamax.
tomk, re copstrat article, worldbeta also cites a critque link. i found it interesting that he uses relative strength to a benchmark and that this presents a better stat than abosolute change. What does not appeal to me is investing in more than one top performer. For each roc strategy I'd rather just have one position. I'd rather get diversity from multiple strategies that each have consistent track records. The best analogy is I rather have one surfer on one wave, but a select group of waves as a means to get more surfers.The diversity will come with mulitiple waves (strategies).Staying invested all the time looks to be prevalent with rotation models, other than a one time exit strategy at a slow moving weekly or monthly average. Timing is too hard, too prone to subjective interpretations. I'm trying to get around that by using cash and zero bonds as asset choices.And, specific tech indicators must confirm end of week. Do you have back testing for the timing part of your model? So far, you seem remarkably in synch.The only strategy that has me in a position rotated to global bonds two weeks ago. Overall, up 1.8% ytd. Russell 3000 is my benchmark.
Better get back in before market makes a new high. I would hate to be short this market gapping up so much. It is great to hear all the pros on here so bearish.
joe
Anon 10:19,
I can't recall if I've ever read other articles on relative strength to a benchmark vs. comparative rates of change. The one thing I dislike about the copstrat strategy is using a single time period (12 months) because it creates too much volatility in the rankings. Ned Davis Research and NoLoad Fund X demonstrated the advantages of using multiple ROCs.
I'm not sure I understand your "one position" strategy comment.
Obviously timing is not necessary to do a relative strength strategy, however, without timing your std dev and drawdowns will be much higher. Keep in mind you will see significant lag before less risky asset classes rise to the top of your ranks.
Timing doesn't have to be hard or subjective, but it does require discipline to work. Imo timing is all about risk/volatility reduction.
I can't formally backtest my timing model because I don't have Goefert's model data. Before I began using Sentimentrader.com I used to track my own sentiment indicators and created models that backtested well. Again, I'm not looking to beat the market through market timing. If my timing model could equal or slightly lag market returns over the long term (at much lower volatility) I'm going to be happy. Outperformance is going to come from selecting superior asset classes via momentum and leverage.
tomk,you've been at this for long time, diligently. I'm persistent because it fits my head and I've done well so far but know how important a simple doable model is. I agree about mulitple periods. For now, instead of taking averages of multiple periods, even with weightings, I just take one and back test for consisent results. The best performer for that strategy is my position...not the top two or three. Then I select another roc period and backtest that one. The best performer is my position,not the top 2-3. For each group of etfs, I want two to three strategies, at least one with a shorter roc and one with a longer roc.Hence, this is how I end up using multiple periods As for timing, I greatly agree with you. Timing alone will not beat the market, it's just to keep me from the big dump...most times, timing will have me retreat relative to my benchmark. Just my way of preserving capital. One will get hurt, though, by not having a plan of re-entry.
Pardon a modest disagreement Anon 8:14 but successfully avoiding 'big dumps' normally makes beating the market (assuming that's what you want to do) much easier to accomplish even if you do struggle with re-entry points. Back in the late 70's and early 80's, the timer's 'bible' was Norman Fosback's (1976) book, "Stock Market Logic." As I recall one of the first 'exercises' was simply to do some basic math -- e.g., if you went down 15% you had to make nearly 18% just to get back to where you started -- this by way of persuading the reader that avoiding serious losses (Roger's "down a lot") was the first (but not the only) step in making serious risk-adjusted profit possible.
rw...no disagreement...that's what I'm trying to do...accept some loss of opporunity, sometimes more than others in order to avoid money loss. p.s. when the indexes hit the 20ema, or a little sooner, for those that want to buy a little insurance as part of a low volatility port, imho, that would be a good time.
Post a Comment