Howdy, Stranger!
It looks like you're new here. If you want to get involved, click one of these buttons!
In this Discussion
-
bee
January 2014
-
cman
January 2014
-
Ted
January 2014
Here's a statement of the obvious: The opinions expressed here are those of the participants, not those of the Mutual Fund Observer. We cannot vouch for the accuracy or appropriateness of any of it, though we do encourage civility and good humor.
Support MFO
Donate through PayPal
Comments
Yesterday was a good day to test if a portfolio was well balanced and guage real beta exposure.
Expect we will zig zag quite a bit this month and test another low this week or next.
Regards,
Ted
One strategy I have tried to execute (using funds or etfs that do not have a transaction fee) is related to a daily rebalancing of winners and losers. Let's say I have two investment that exhibit low correlation to one another (they often move in opposite directions on certain days). In an attempt to tamp down the daily volatility I exchange a small percentage between the two investments so long as they exhibit opposite perfomance (one up...the other down).
Recently both QQQ and GDX have exhibited upward momentum, but often on different days. Today, for example, QQQ was up almost 2% and GDX was down over 2%. I rebalance the sum of the difference between the two. So today was a 4% sell of QQQ and the proceeds where used to buy GDX. If they move in harmony (up together or down together) I stay put.
Not sure if this will reduce volatility or provide any other benefit. Criticism welcome...
It is useful to look at the assumption behind rebalancing with low correlation assets. First, it is low correlation not inverse correlation with the assumption that both of them will go up in time but in different cycles. So, for example, you cannot do this with long and short etfs say QLD and QID. But your example of QQQ and GDX would satisfy that.
Even then, I believe there are studies that show rebalancing hurts performance than help in the average but the reason you do that is to manage volatility/risk as things get out of balance. More frequent the rebalancing, the worse the performance hit. The reason for this is that very often the rebalancing becomes throwing good money after bad as the funds diverge and they can remain divergent for a long time. In other words, you are not giving enough chance for the money to exploit the currently trending asset. Look at the long period in which QQQ kept going up and GDX kept going down over months this year. You would have transfered a lot of your money to GDX quickly while this divergence was still going on and losing monet in GDX and would not have benefitted much from QQQ. If the algorithm didn't transfer a lot then the effect of that transfer would be in the noise relative to just holding both.
You can look at history and come up with a sequence of trades that would optimize for what happened in that period, but I don't think there is a mechanistic formula that works for any sequence as may happen in the future. You can look at your strategy and trend/momentum strategies as extremes in a spectrum. The latter rotates when the trend starts to reverse while the former preemptively rotates in hope of a reversal soon. If there are very frequent reversals, your strategy will do better than trend following and vice versa. But there seems to be more persistence in trends than frequent reversals over time for common asset classes.
Perhaps your strategy works better for highly volatile assets, say GLL and AGQ. But I suspect it will still destroy performance over time unless you get lucky.
Movements over a day or two neither validates/invalidates sector leadership nor create/destroy momentum/trend, so I don't see much practical use unless one is trading and trying to time it perfectly on a daily basis. Some people try.
I guess I am trying to figure out ways to optimism the positive attributes of volatilty. Probably just a bit frustrated from looking at all of this a little too closely. Thanks for you detailed response.
Here's another example of where I have also thought that optimizing volatilty might add some profitablilty.
Take YACKX, a steady LC fund and CAMAX, a much more volatile "opportunistic" LC fund. By charting these two funds together I believe an investor could optimize CAMAX's volatility as it is referenced by YACKX performance. When CAMAX periodically outperforms reallocate some CAMAX into YACKX. When it underperfroms move some of YACKX into CAMAX.
Here's a recent three month chart illustrating a 10% outperformance by CAMAX relative to YACKX. This would trigger a reallocation.
Isn't what you are pointing to in this case simply taking some profits from one fund where the other asset was could have been cash? This is what happens in my play money portfolio with leveraged funds. That would be prudent if you invest in volatile assets anyway. Whether you reinvest in another fund is an independent decision. So your strategy is equivalent to "take regular profits"? Which is fine if you have a good entry and exit strategy for your buys and sells.