The progress of the cybernetic system for the two unleveraged ETF portfolios is meeting my expectations. I was aiming for 15% annual return with <10% maximum drawdown. Both portfolios are a little below 15% with drawdowns substantially below 10% (see here and here). However, since all positions are benefiting from QE2 courtesy of Helicopter Ben, I expect some losses due to either a lull in debt monetization or its end (which I consider to be unlikely) which will cause a pullback or reversal respectively in asset prices. A reversal would be reminiscent of the fall of 2008, so it will be interesting to see if my system will behave as it did when I backtested it.
I have been wondering about how leverage would effect my system. There are three methods of adding leverage. The first is to use margin with the current portfolios. Since interest rates are currently so low, this is a viable idea. For instance, take a look at margin rates offered by Interactive Brokers (disclosure: I am a customer of IB). However, when interest rates rise (for those predicting rates to skyrocket in the near future, I remind them of the curious case of the last 15 years of the Japanese government bond market) this strategy will no longer be viable (think of the margin rates in the late 70s/early 80s). The second method is to use leveraged ETFs. This strategy suffers from the following problems: tracking errors, short histories, and replication (i.e. finding a leveraged broad based commodities index). The third method is to use futures. This is the most cost effective and efficient way to obtain leverage but it also requires a larger account size than the other methods due to risk management issues.
I wanted to construct a futures portfolio that was in the spirit of the small ETF portfolio. It should minimize the number of holdings (while obeying my asset allocation rules) thus making it easy to trade. Additionally, it should be accessible to a non-hedge fund audience (meaning that large account sizes in the hundreds of thousands are not needed). Here is the portfolio:
1. US equities: ES – emini SP500
2. International equities: ER – emini MSCI EMI
3. Currencies: YG (emini 1/3) or MGC (emicro 1/10) gold
4. US treasuries: ZN – 10 year note
a) QM (emini 1/2) crude oil
b) XK (emini 1/5) soybeans
Since the S&P 500 index comprises multinational companies, it provides exposure to developed equity markets and the US equity market, so I use an emerging market index for international equities. The 10 year note is a proxy for the entire yield curve. Gold is a proxy for all (fiat) currencies (the US dollar index only measures the US$ vs a basket of currencies). For commodities, I made a decision to only include two contracts, fully realizing the limitations of this approach. However, I was determined to keep the number of contracts as small as possible, so I believe that this is a reasonable compromise. I chose crude oil and food over base metals and other components of a commodities index. Since soybeans are used in animal feed and many food products, I use it as a proxy for all food products. Note that there is no liquid futures contract for real estate.
To trade this portfolio, I will use rules very similar to those used for the ETF portfolios. The initial trades will require some thought. Some of these markets are so overbought (I am fully cognizant of the fact that this is an imprecise term) that I will not enter them because I would need a very large portfolio in order to obey my risk management rules. So I will have to wait until my risk management rules permit me to enter some markets. For ETFs, trading decisions were only made at the end of each month. Due to rollover, I will have to make new trades more frequently. Initial trades will be made on Monday, Feb. 28th, and will be posted to this blog.
Note that I will no longer be trading the small ETF portfolio on collective2. I will be trading this new futures portfolio instead. I will also continue to trade the ETF hedge fund portfolio on collective2.