Rotational Portfolio Hates This Market
Rotational combines component rotation and asset class rotation to hold a small basket of ETFs or ETNs, selecting the handful with the most momentum from a representative sampling of classes and components. Throughout this article, when I refer to momentum, I am referring to an exponentially smoothed measure based solely on price movement.
Information is as of the close on July 25, 2008.
Model Allocation
Based on beginning with a $100,000 portfolio at inception, these are the current weights and holdings. The initial target was a buy of 10% weights per position. See my previous post on this system. Sort is alpha order by ticker and weights are rounded to the tenth of a percent.
Brazil (EWZ) 10.2% weight
Oil Equip/Srvcs (IEZ) 10% weight
Natural Resources (IGE) 9% weight
Agricultural (MOO) 8.6% weight
Oil Services (OIH) 9.7% weight
Energy Exploration (PXE) 8.9% weight
Steel (SLX) 9.4% weight
Natural Gas (UNG) 9.8% weight
Oil (USO) 14.8% weight
Materials (XME) 9.7% weight
Cash -0.1% weight
Returns
Based on beginning with a $100,000 portfolio at inception.
Equity: $89,942.44
Gain, Past 4 Weeks: -17.09%
Gain, Year to Date: -11.49%
Gain, Since Inception on 11/19/2007: -10.06%
None of the ETFs in the Rotational portfolio paid dividends or distributions in the past four weeks.
Total dividends = $0.00 on the tracking portfolio. This amount is included in the returns shown above, and will remain in cash until needed for a new purchase. Note, commissions are expensed at $10.00 per trade when accounting for returns.
Changes To Model Allocation
Rotational screens for momentum inside a list of ETFs and ETNs by asset class category. The system is holding the top 10 issues, ranked by momentum, regardless of which asset class they are in or how much momentum they have.
If this system were to be initiated today, the target allocation would be a buy for 10% weight holdings of the ten issues highlighted in gold or green in the table below. Items highlighted in gray are “sells” from the existing model portfolio.
If the table is truncated in your browser, click on it to view it in its own pane. Depending on your browser, you may have to click again to view it in full size.
Tracking
Shares of EWZ, MOO, and SLX will be sold, market at open on Monday. The proceeds, plus cash, comprise 28.1% of portfolio weight, and will be used to buy shares of DBA, GLD, and SLV based on the closing prices on July 25, at 9.4% weight each. I will round down any fractions in the share calculation.
Commentary
Below, I present the change in rotational momentum from the last evaluation to the current one. It can be quite instructive.
Here is a table that shows the average momentum for the different issues in each asset class, at different evaluation dates from the inception of the program.
Bonds, as an asset class on average, are still in negative momentum. They’ve actually fallen a bit further since the last update. no longer have some positive momentum. The biggest droppers in momentum are the emerging market bonds, followed by the junk. The short end of the Treasuries (1-3 years) haven’t really changed in momentum over the last four weeks, and are still sitting at close to zero in my measurement timeframe. This after darn near a year of positive momentum from the short end.
Commodities as a class still have the most momentum, but are still losing momentum rapidly. This story hasn’t changed in about 16 weeks - which points out why I only update once every four weeks, because the markets just don’t change that quickly. Or do they? It’s been a stupendous implosion in Natural Gas, which unfortunately the portfolio has been long. While precious metals have been gainers, they haven’t been able to overcome the “great sucking sound” coming from energy. This represents a possible resurgence of “inflation fear.”
Currencies competing against the dollar stabilized in momentum, actually generating a small increase in momentum, so small I would consider it practically “no change.” The British Pound and the “carry trade” tracker DBV lead the gainers in momentum over the past four weeks, while the Yen and Swedish Krona are falling in momentum. It’s interesting that, in the midst of a strong move down in equities and risk-taking, the “carry trade” is still steady.
The foreign stock markets fell out of the turnip truck last month, and THIS month, the turnip truck shifted into reverse and just frickin’ ran over the foreign stock markets. As an equal-weighted group, this is the worst momentum the foreign markets have had since I’ve been tracking this system. ALL of the foreign markets lost momentum over the last four weeks. Russia and Brazil are the strongest foreign markets.
The domestic industry groups fell off of the turnip truck this month, with negative momentum pretty darn close to that they had at the March lows. The only gainers were gold miners, biotech, pharma, and health care equipment stocks. The big industry losers were the high-momentum gainers of the last several months, groups like steel, basic materials, and the oil/energy complex. The technology winners from the previous update have fallen by the wayside.
REITs were punished just as badly as the domestic industry groups were, and if not for the incredibly poor momentum held by foreign markets, REITs would be the worst-looking class right now. All classes of REITs fell, although residential and industrial/office REITs continue to hang around, falling the least over the last month and still having the best momentum outlook.
This has remained a tough market to call, and while it hadn’t seemed to bother Rotational before last month, it’s bothering the portfolio NOW. The portfolio has been heavy in materials and energy, which have taken a terrific pounding over the last four weeks. Rotational is a trend-following system that believes there’s always a bull market somewhere, and in previous downturns this year, the portfolio has been able to ride it out OK because it was in “what was working.” Not this time. The system doesn’t change, however. It just executes, regardless of circumstance, and while it does tend to to suck wind at turning points, it makes up for it in the straightaways - because there’s almost always something trending.
If you’d like to become of member of The Rempel Report, you can register here. Members receive email notification of new posts and can contribute to the site through comments. Registration is still free!
July 27th, 2008 at 9:07 pm
The rotational system strikes me as one with fairly significant drawdowns, which likely occur during the aforementioned turning points as well as times when “momentum” stocks as a group are out of favor. Has there been an effort to minimize drawdowns by detecting those turning points and momentum-hating periods, and staying in cash during those times?
July 27th, 2008 at 9:28 pm
Don, thanks for commenting.
The point about momentum stocks is off-target for this system, because the system can follow commodities, foreign currencies, bonds, domestic industry groups, REITs, or international indices, as need be. During times that traditional momentum industries (say, Tech) are out of favor, SOMETHING else will be IN favor. Example, during the “bear” market of 2000-2002/3, it wasn’t much of a bear for those in Treasuries, gold and gold miners, etc.
It’s theoretically possible to find some keys to reduce drawdown at turning points, and some ideas I’m investigating are (1) stop losses, (2) using the relative strength spread from worst to first, and (3) keeping a dedicated short position. So far, I haven’t been able to reduce drawdown in backtest without deleteriously impacting return, past what I’m happy with. Oh, bother.
Working on the systems is a continuing project.
July 28th, 2008 at 3:20 am
Bill,
I’m sure you have considered this.
Add a requirement that the momentum signal (say PPO) should be above its 21 day EMA for both purchase and retention of the asset.
Selection remains with high PPO subject to the above.
Currently, top 6 picks would be eliminated.
I still wish you were following the “forced allocation” system to ensure some diversification. Maybe allow a maximum of 2 picks from any one class?
Steve
July 28th, 2008 at 3:43 am
An interesting article for me.
I have noted in my own trading since the beginning of May that the momentum has really dropped off. My average hold time for NYSE/NASDAQ stocks is only 2.5 hours but all of the stats I keep track of indicate that not only is there less follow through there are more reversals.
My current belief is that this market which is quite a tough one for momentum players should be a great one for people who look to range trade, fade breakouts and momentum.
Haven’t been able to translate that into a working system quite yet, but am very interested in tracking what others are experiencing, who and what strategies are doing well. I think that a skilled trader can paddle against the current but it’s a much simpler and potentially more lucrative idea to find which direction the current is heading and set sail accordingly.
Andrew.
July 28th, 2008 at 4:54 am
Steveal, performance in backtest is actually worse using a short-term moving average filter. You get slammed by selling near the bottom of many minor moves, even though you would be out of the few major swatdowns.
Everything’s there for someone else to use the research and follow or track a “forced allocation” system, using the charts above.
August 3rd, 2008 at 8:45 am
Top ten are now (alpha order) DBA, GLD, IBB, IEZ, OIH, PXE, SLV, UNG, USO, and XME.
What’s amazing is how LOW the cutoff point for top ten momentum is, compared to where it’s been in the past. I suspect that at the next full update, we’ll have a decidedly different set of leaders than we did at the last full update.
August 9th, 2008 at 12:29 pm
The top ten are now (alpha order) FXA, FXE, FXM, GLD, IBB, IEZ, OIH, SLV, USO, and XME, and the cutoff point for the top ten is even LOWER. The leadership is changing rapidly, and many of the classes I track are deep in negative momentum territory.
August 17th, 2008 at 8:31 am
The trend continues, with turnover in the top ten, and with the cutoff point being even lower. I doubt the process of finding new, lasting leadership will be completed by the next update for this system, which probably means it will continue to suffer for a few more weeks.
Alpha order, leaders are now: EVX, FXE, FXM, IBB, IEZ, IHI, IYT, PHO, SWH, and USO.