Is this month going to be an example of DM investors getting whipsawed? What price would the S&P500 have to close at on October 1 to cause you to sell out of treasuries/cash and buy back into the market? What would your effective losses (missed appreciation plus transaction costs plus taxes) be on having followed a DM strategy in this case?
Conversely, what price would the S&P500 have to close at on October 1 for you to have profited, in this instance, on following a DM strategy? Is it just the Sept 1 price of 1913.95 plus frictional costs plus the 1 month of returns on whatever other asset (treasuries/cash?) you moved into instead?
The month isn't over yet and volatility is so high these days that I won't pretend to know what the October 1 price will actually be. But watching this strategy unfold in real time in this thread has highlighted for me that every asset class swap a DM trader makes is essentially a bet against the 30-day future price of the asset they sold out of, relative to the asset they bought into, minus any trading costs and/or taxes. If every trade is just a bet against the 30-day future price, how does the strategy compare to just buying put options?
Every strategy is going to have good and bad months, and I'm not posing these questions to pick on one. Whipsaws are a recognized risk of a DM strategy, an accepted infrequent loss that you swallow in exchange for the downside protection you expect to get most of the time. I'm just trying to look forward a little, to establish expectations for how this strategy will perform compared to a buy and hold strategy depending on what the market might do in the short term.
If I understand correctly, a DM trader who swaps out of stocks, as some people here seem to have done, is essentially betting that the short term price change of the stock market will continue to move down until it is below some new lower specific price determined by the frictional trading costs and the expected return of the alternative investment, which is typically well known for 30 day treasuries or cash. If they are correct, they will have avoided a known amount of paper losses but maybe incurred some transaction costs. If they are incorrect, they will have incurrred a known amount of paper losses, plus maybe some transaction costs. I'm just not clear on the relative quantities of the losses in those two scenarios and how it relates to the real future market price relative to the price they're betting it will fall below.
As with the one month performance any active strategy, compared to the S&P this month will be an example of
1. A convincing outperformance of dual momentum,
2. A convincing underperformance of dual momentum,
Or.
3. A rough equivalence of DM,
Which of the three is anybody's guess (aside from those who can predict the future.)
A whipsaw it will not be, as that would require a change in direction at the time of the next position switch.
The required action at the close of 9/30 is dependent on the future returns of short term treasuries, the future returns of the S&P, and the future returns of EFA, none of which are knowable. I can tell you what I would do today if today were my trade day, but that's about it.
Based on your comment, I am once again left with the impression That you don't really understand the strategy Sol, the DM Practitioner is simply betting that absolute momentum will give him/her a reproducible signal for the beginning of a bear market. He/she trades about an 80%probability of this signal being correct , for the knowledge that 20% of the time he will miss out on some upside. It's a trade of big downside for small upside. And it's a trade that I'm happy to make again and again regardless of what happens for the rest of the month.
If you want a feel for the history of the strategy and how frequently it loses to the broader market on a monthly basis, feel free to do some homework and look through the wealth of data at optimalmomentum.com
Here's a link.
http://www.optimalmomentum.com/trackrecord2.htmlAn actual interesting facet of the current market situation is just how flat the market was for the first eight months of the year prior to the recent volatility. This essentially means that look back periods from 3 to 8 months are all equivalent in terms of their signal.