I can hardly believe it but I understand that some of my readers are not daytraders at all but actually have jobs and otherwise spend their days not glued to multiple monitors watching 2 minute bars flow by.I can almost recall a similar lifestyle many years ago, but being old my memory is a bit faded so I'll just have to take it on vague faith that I too once enjoyed that leisurely market approach.
The braintrust at ETF Rewind has graciously provided me with a variety of prospecting tools that have recently permitted me to fantasize about a return to those simpler trading days when high frequency trading, prop shop manipulation, algo program trading and dark pool shenanigans didn't make swing trading the equivalent of navigating a mine field while blindfolded.
So . .as we approach the new year here's a simple rotational system based on a 10 period momentum model with a simple PSAR stop that trades only 4 ETFs . . QLD (2x bullish QQQQ), EEM (emerging markets), DBC (2x commodity index) and VTI(vangauard total market).
The system is based on weekly bars and readjusts every Friday based on the momentum signal.
If momentum is trending up the model invests in the strongest of QLD, EEM or DBC.
If momentum is trending down the model sells short the VTI.
That's it.
Performance metrics are shown relative to a buy and hold SPY position.
While not currently available for subscription on the Rewind site, I expect this system and variations (currency, sector, country and volatility rotational models) to become available in the new year along with other cutting edge prospecting enhancements. Then I can relax, spend my time paddling around in the pool, golfing and otherwise enjoying a stress-free trading income.
Can't wait.
12 comments:
A simpler variation of this model will be posted each week for free here: http://marketrewind.blogspot.com/2009/12/mrkt-rotation-model.html
Merry Christmas, Jeff
How did this do during the last year compared with the SPY? May want to use RWM to double test that VTI short. Is there a total market inverse for retirement accounts? Best, J
OK, this model has been added, Best J
http://marketrewind.blogspot.com/2009/12/bzb-traders-rotation-model-added.html
Nice model.
The momentum to determine up or down is based on the SPY? How do you identify which one is the "strongest of QLD, EEM or DBC"? And do you use the standard parameter for the PSAR?
Thanks!
Thomas,
The momentum algorithm is the property of ETF Rewind and Jeff has revealed the indicator that drives the rotation model in this weekend's update for subscribers. Consider a subscription or at least a free trial to see what's what. I'm working on my own rotation model using the same 4ETFs but driven by my 3 finger lead and reverse algorithms. My goal is to reduce the "weeks to recover" number to less than 3. This may involve a fractal model that actually readjusts every 3 days instead of every week. Just another project to keep my busy.
does PSAR stand for a parabolic stop?
thanks
jt,
Correct. Typical SAR settings are .02/.2, but I've found best results are obtained when optimized in relation to the beta of the underlying. Higher the beta, shorter the step.
Hey, I wish you the best of New Years, and hope you do Lazy Out!
On a side note, I wish you would stop changing your picture, every time you do, it makes me feel older and older, HAHAHAHAHAHA!
Good stuff as usual... I have a couple of basic questions in how you trade this.
My understanding is that you would look at the model every Friday, go long the highest-rated ETF, and close out whatever other position you may have had. When you go long, you set a stop in case momentum changes during the week. What I'm not sure about is how a PSAR stop can be implemented. Is this something that your trading platform provides? Is it likely that Ameritrade offers somthing similar?
Thanks...
Ty,
I'm afraid I don't know anything about Ameritrade's programming abilities.
Basically the PSAR acts as a go/no go check for the Friday entry. If the parabolic signal is in conflict with the program BUY signal, the program goes to cash and waits for the next week's signal. There is no intra-week stop. Jeff and I are still working out the kinks in this rotational model and intend to integrate an equity curve stop (and others) into future iterations.
Interesting model. I also added a price above moving average entry requirement which improved my backtesting somewhat.
Since QLD has been the choice for many weeks now and is likely to be so in the near future, would you scale in now or wait for the next change?
Thanks in advance,
Gary
Gary,
Apologize for the delay. I've been overwhelmed with emails lately.
The model is strictly rotational, so no add-to.
We're adding several filters to the model in an effort to forecast likely highest % gainers. Hopefully that will be in place in a week or two.
Post a Comment