January 2014 Performance Review
Note that all results are based on computer-generated trading signals and not actual trades. Cost of $8 per side for stocks and ETFs, and $8 per side for futures, have been deducted. Historic performance is simulated. More detail about these strategies can be found on the website. Neither actual nor simulated profits are indicative of future results. See full performance table at the end of this document.
The stock market started the year on a sour note, with all three of the benchmark indexes off, and the SPY off the most. Of our 15 stock portfolios, only 6 were net profitable in January, but all were better than the SPY. Futures strategies were more difficult, with the NewEdge CTA index down 2.30% for January and the Trend index down 4.45%. Of our nine futures programs, only two beat the benchmark. We’ll take a closer look at the Futures market breakdown later in this report to show how patterns are changing.
We are pleased to announce that we’ve added the Weekly Futures to the Trend Program, now indicated as Group 7 (G7). As with the weekly stocks (G6) we expect less activity and somewhat smoother performance using weekly data.
Note that the “All Signals” reports show short sales in stocks and ETFs, even though shorts are not taken in the portfolios. We are currently studying using the inverse ETFs in the Trend and Divergence programs in order to offer greater diversification and the ability to profit during a declining period. We will notify you when we have results.
PEFORMANCE BY GROUP
NOTE that the charts show below represent performance “tracking,” that is, the older results are simulated but the new returns are the systematic daily performance added day by day. Any changes to the strategies do not affect the past performance.
Group 1 Trend Program for Stocks and ETFs
The Trend program seeks long-term directional changes in markets and chooses stocks that are consistent over many years combined with good short-term term returns. The Trend portfolios for stocks and ETFs all outperformed its benchmark index SPY in January, although only the stock portfolios showed a net gain. The charts below show that all five programs are still in a strong profit pattern, although the Rotation program has a more volatile, sideways pattern in January. When selecting ETFs for the portfolio, we must expect them to track closer to the broader index, while selecting individual stocks can allow the system to find uptrends in an otherwise poor market. We need to remember that the stock portfolios choose high beta markets, so larger drawdowns may alternate with larger gains.
Group 2 Divergence Program for Stocks and ETFs
The Divergence program looks for patterns where price and momentum diverge, then takes a position in anticipation of the pattern resolving itself in a predictable direction, most often in the way price had moved before the period of uncertainty. January posted mixed results, with the smaller portfolio of 10 stocks showing positive returns, while the larger 30-stock portfolio had a small loss. The ETF program had a slightly larger loss, but less than the benchmark, because many of the ETFs track the overall market indexes closely. The performance in the chart below shows that returns remain in an uptrend. While the smaller 10-stock portfolio is outperforming the others in returns, it is also more volatile because of the fewer stocks.
Group 3 Timing Program for Stocks and ETF Rotation
The Timing program is a relative value arbitrage, taking advantage of undervalued stocks relative to its index. Its primary advantage is that it doesn’t depend on the market direction for profits, although these portfolios are long-only because they are most often used in retirement accounts.
One of the problems with buying undervalued stocks is that they can continue to get even more undervalued. In addition, the overall market can turn down so that even a relative gain over the benchmark SPY can actually be a net loss. To avoid those situations, the program has an absolute stop loss, plus it hedges the overall risk of the portfolio in three parts if the broad index turns down. Given the long rise in stock prices, a correction is inevitable.
In January, all three Timing portfolios beat the benchmark, with the smallest 10-stock portfolio showing a net gain. Nevertheless, it is important to be prepared for a downturn in the overall market. The charts below show the historic performance of the three Timing portfolios with and without the SPY hedge. As a reminder, the hedge is placed in three parts based on three different trends. The size of the hedge is determined by the combined volatility of prices of all positions in the portfolio. The maximum hedge is 50% of the total risk, so that each of the three hedges account for 1/6 of the risk. The reason for this is that the hedge rarely nets a profit. Most often we enter a short SPY position (or long SDS) and the market turns back up. Of course, that’s preferable rather than continuing lower. We persist with using the hedge because of the rare year, such as 2008, where prices continued lower. With the hedge, losses were reduced by about 40%. We feel that taking a smaller profit in most years is a good trade-off when very bad years show much smaller losses.
To allow you to judge for yourself, the two charts below show returns with and without the hedge. You’ll see that the combined performance, on the left, is clearly smoother for the larger, 30-stock portfolio, the best indication. The smaller 10-stock portfolio will always be more volatile, but less so with the hedge. Given the increasing likelihood of a reversal in stocks, we recommend using the hedge.
Group 4 Trend Program for Futures
Futures allow both high leverage and true diversification. The larger portfolios, such as the $1 million shown in the G4 chart below (the green line), are diversified into both commodities and world index and interest rate markets, in addition to foreign exchange. Its performance is not expected to track the U.S. stock market and is a hedge in every sense because it is uncorrelated. As the portfolio becomes more diversified its returns are more stable.
The leverage available in futures markets allows us to manage the risk in the portfolio, something not possible to the same degree with stocks. This portfolio targets 14% volatility. Investors interesting in lower leverage can simply scale all positions equally in proportion to their volatility preference.
January was not a good month for our Trend futures program or for the CTA industry in general. NewEdge shows that all CTA programs lost an average of 2.30% and trend programs lost an average of 4.45%. Our portfolios did slightly worse, but then we try to balance the risk across sectors as equally as possible. We target 14% volatility, so that if the industry has a lower average we would post both larger losses and large gains. The Trend Futures portfolios lost 2.86%, 6.68%, and 9.00% for the $100K, $500K, and $1 million portfolios, respectively.
Looking closely at the interest rate sectors, which have traditionally been the source of the largest returns, we see a very different pattern in the short rates and bonds. All trends are now up again, indicating a move to lower yields. If we compare the short end, using Eurodollars, and the long end, with U.S. 30-year bonds, we get the pattern:
- Eurodollars have been long all through January having had cut its long position to ½ on December 19. As of January 31 it has reset a full long position
- U.S. bonds were short going into January and only reversed to a small long position on January 24.
We can see from the results below that being long was the better choice. Still, we trade a balanced portfolio to gain diversification and offset risk.
In the charts about, the CSI codes are TU=2-year notes, ED=Eurodollars, FSS=Short sterling, FEI=Euribor, EBS=Schatz, US=30-year bonds, TY=10-year notes, FV=5-year notes, EBL=Eurobund, EBM=Eurobobl.
Note that there have been small changes to our largest portfolio, replacing NYMEX crude oil with London crude oil and adding gold to the commodities sector.
Group 5 Divergence Program for Futures
This short-term pattern recognition program has the highest long-term information ratio and very consistent returns for the past 20 years. The G5 NAV chart shows the largest $500K account outperforming the others, but at the same time shows greater inconsistency. The largest, most diversified $1 million account (in green) is clearly the most stable but with lower overall returns, consistent with expectations. The smaller accounts may vary based on the impact of particular markets in any one month.
January performance was slightly worse than the benchmark NewEdge CTA returns, with the $100K, $500K, and $1 million portfolios losing 2.86%, 6.68%, and 9.00%, respectively. The overall performance remains very positive.
As with the Trend Futures program (G4) we’ll look at the way the interest rate markets performed in January. In the chart below the short-term rates are on the left and the longer maturities on the right. It is interesting that the performance pattern is opposite for these two sectors and the reverse of the performance of the Trend program, which had gains for the short end and losses for the bonds. As you can see, these two groups came close to offsetting one another. Most programs count on profits in these two sectors to carry the rest of the portfolio during difficult market periods.
Group 6 Weekly Trend Program for Stocks and ETFs
We’ve said that the Weekly Trend programs are not just for lazy investors, but they serve an important purpose of their own – they smooth out the data by ignoring the mid-week noise. That makes sense for long-term trend following but not for the other short-term programs. The performance table at the beginning of this report shows that the Weekly Stock and ETF Program outperforms its daily counterpart in all ways, returns and information ratio. The only problem is psychological – the need to wait for the end of the week to change you positions even when the market is going down.
A close study of markets shows that stocks, compared to interest rates, currencies, and other futures markets, are far noisier. It may seem hard to believe because stocks have gone straight up for the past three years, but it’s true. The best sector for trend returns is interest rates and the worst is usually stocks. The reason why we diversify is that we can never know when a run such as the past three years will occur, and we can’t afford to miss it.
In January the Weekly Trend Stock and ETF Program was the best performer, netting a gain of 6.22% in the 10-stock portfolio and beating the daily Trend Program in 3 of 4 comparable portfolios. The charts below show the weekly NAVs and are very similar to the performance of the daily program.
Group 7 Weekly Trend Program for Futures
This program is new and is essentially the weekly version of Group 4. There were minor changes in the $1 million portfolio which were also made in the daily portfolio so that these programs are consistent.
Over the long term, this weekly program has a higher information ratio than the daily version, but it performed slightly worse in two of the three portfolios in January, netting losses of 4.61%, 7.64%, and 8.29% for the $100K, $500K, and $1 million portfolios, respectively.
In the chart below, all three portfolios show greater consistency than the daily version. Still, there is a period of two years for the $100K and $500K portfolios where returns were sideways to negative and the most recent month shows added volatility, although well within the normal range. In order to trade a weekly program an investor must have a long-term view of performance and an understanding that changing from long to short during the week isn’t always going to net a larger profit. More often than not, it’s a false signal and can be avoided using a weekly program.
KaufmanSignals.com February 2014