Our programs saw two exceptionally large price shocks during January, the Swiss franc in futures midmonth, and Hawaiian Holdings (HA) on the last day of the month. That makes it a good time to discuss these ugly, but not so unusual events. What can we do to reduce the loss? And why do they always go against us?
First let’s look at some information about price shocks. We’re going to identify a shock as a day in which the high or low is more than 3.5 ATRs (Average True Range) above the average ATR, so if the normal range of AAPL is about $3 (trading at $115), then we need an intraday move of $10.50 to trigger a “shock.” We did the same for SPY, using a factor of 3.0. SPY would be less volatile because it’s the average of many stocks. The charts below show the most extreme move of the day, in percentage, and the closing percentage change on the same day. Many of the shocks in AAPL seem to revert during the day, while not as much for SPY. We would venture to say that a shock in SPY is one affecting the whole market, and may be more structural, such as poor overall economic news. Still, when averaged, a typical pullback from the extreme to the close is 28% for AAPL and 24% for SPY.
Since 2000, about 15 years, there have been 40-50 price shocks in each market, that’s about 3 per year. Apple shows 6 over 15%, although when we think of the spectacular rise in AAPL prices, we don’t remember those days. New highs seem to make past risks less important.
Why do most price shocks cause losses for the majority of investors? Because we are mostly long and there is a void of liquidity selling on unexpected news. When good news is a surprise, the price moves far less.
What can be done to reduce the loss on those days? That’s a tricky question. By definition, a price shock is unexpected, so studying old price data is, for the most part, meaningless. Can we use a stop-loss? Stops work if the price keeps going, but experience tells us that we normally get out at a terrible price, then we see some recovery by the close, so if you’re going to exit, then the close is better than an intraday stop order. But then, if you’re a trend follower and have a very large profit in AAPL or HA, and a price shock takes away 10% in a day, is that important? If you get out, but the trend is still up, you might miss the next 50% gain. How do you decide when to get back in? The best solution is to have a system that includes risk controls, which could even be a simple moving average where you exit when the trend changes, and follow those rules.
Rather than doubling down and hoping for a reversal, there are two other ways to reduce the loss of a price shock. Naturally, the first is diversification. If one stock drops 20% and it’s only one of the 20 stocks that your trade, then the net effect is a loss of 1%. That’s a lot better. The other is to have a trading strategy that is out of the market more often. If your method is only exposed 25% of the time, you have a 75% chance of missing the price shock. We see those as the best choices.
Trend Strength Index
One measure of market strength is our Trend Strength Index. Our Trend strategy is a composite of many trends, medium term to slow applied to about 250 stocks. When combined, these determine the position size of the current trade. If the faster trends are down but the slower one up, then the position size might be zero. The appearance is that trend positions scale in and out based on the strength of the trend. The Trend Strength Index appears at the bottom of the Trend Stocks All Signals report each day. We’ve tracked it from the beginning of 2014, and the chart below compares it with the SPY. TSI is the Trend Strength Index and SPY is the SPDR ETF. TSI values about zero indicate a positive trend. The range of the TSI is +1 to -1.
We could just reprint what was said last month, showing underlying weakness in the stocks even while prices inch higher. But this month they seem to have stalled and volatility has increased. Volatility can scare away investors, especially when the stock market is up one day and down the next. If we look only at the Trend Strength Index we see weakness. But even with the drop in the TSI last July, and again in October, the SPY recovered quickly. The U.S. economic picture looks good from the statistics, but there are underlying issues in Europe, Asia, and South America that undermine our strength. Given that mixed scenario, we’re going to predict the future as sideways with somewhat more volatility – what we call a good “traders market.”
We offer this Index for those investors who select their own trades rather than following our sample portfolios. Daily Index values are available to subscribers.
Overview of January
Even with some of the portfolios absorbing large negative price shocks in January, 22 of 24 programs beat the SPY and 16 of 24 were profitable. The worst performance was the 10 Stock Trend program, which turned from a profit to a loss on the last day of January due to a long in HA. The best program was the small Trend Futures, which gained yet another 18%. The Equity Trend program had mixed results, with the larger 30 stock portfolio absorbing the loss from HA far better than the smaller 10 stock portfolio and beating the SPY. All three ETF portfolios posted gains.
The Divergence Equity Program was also mixed, with the fractional losses in the 10 stock and 8 ETF portfolios, and a 1% gain in the larger 30 stock portfolio. The Timing Program is now 2/3 hedged in both the SPY and QQQ against long positions in equities. That’s as negative on the stock market as we’ve seen in more than a year. Monthly performance was fractionally lower.
Futures were much better than equities, continuing their long run of profits. The price shock caused by the Swiss Central Bank removing currency limits ironically affected the bigger Trend portfolios, which are more diversified. Returns ranged from +18% to breakeven. The Divergence Futures portfolio had no positions in either market and posted modest gains. More about this in the Futures section.
The major index ETFs, shown in the chart below, look weaker than last month. While the U.S. economy has the perception of being strong, workers are earning less and the housing market, a strong driver of the economy, remains sluggish. The world has gotten more complex and globalization means that investors are watching Europe and its renewed struggle with Greece, as well as China and India, both filled with uncertainty. We expect more sideways price movement with greater volatility this year.
Strongest and Weakest Sectors
There are two ways to view sector rotation, trade the strongest expecting them to stay strong, or trade the weakest expecting the business cycle to rotate them to the top. We have both. The Trend Rotation trades the strongest and the Timing Rotation trades the weakest. The Trend program may hold positions for a long time, so it’s possible for two ETFs to be in both programs. For example, XOP (Oil and Gas) can be in a long-term uptrend, but a short-term oversold situation.
The Trend Sector ETF program buys the 6 strongest sectors of the SPDRs. This month started with Preferred stocks (PFF), Heath Care (XLV), Staples (XLP), Utilities (XLU), Vanguard Reits (VNQ) and Retail (XRT) and on Friday, Jan 31, placed orders to remove XLP (Staples) and XLV (Health Care) with no orders to add new ETFs. That means the program cannot find any ETFs with a performance history good enough to trade at this time, an indication of a weakening market. Current positions are:
Preferred stocks (PFF), Utilities (XLU), Vanguard Reits (VNQ) and Retail (XRT)
The Timing Rotation program (buy low) exited all of its previous positions during January: Preferred (PFF), Energy (XLE), Technology (XLK), Staples (XLP), and Health Care (XLV). It is now holding:
Financials (XLF) and Industrials (XLI), with a 2/3 hedge in SPY based on the combined volatility of the two ETFs.
A Standing Note on Short Sales
Note that the “All Signals” reports show short sales in stocks and ETFs, even though short positions are not executed in the portfolios. Our review of using inverse ETFs to hedge stocks during a decline showed that downturns in the stock market are most often short-lived and it is difficult to capture those moves with trend systems. This confirms our approach to the Timing systems, which hedges up to 50% of the long stock risk using multiple trends. In the long run, returns from the hedges are net losses; however, during 2008 the gains were welcomed and reduced losses. In any correction we prefer paying for risk insurance, even without the expectation of a net gain.
Portfolio Methodology in Brief
All of the programs, stocks, ETFs, and futures, use the same basic portfolio technology. They all exploit the persistence of performance, that is, they seek those markets with good long-term and short-term returns, rank them, then choose the best, subject to liquidity, an existing current signal, with limitations on how many can be chosen from each sector. If there are not enough stocks or futures markets that satisfy all the conditions, then the portfolio holds fewer assets. In general, these portfolios are high beta, showing higher returns and higher risk, but have had a history of consistently out-performing the broad market index in all traditional measures.
PEFORMANCE BY GROUP
The charts show below represent performance “tracking,” that is, the oldest results are simulated but the newer returns are the systematic daily performance added day by day. Any changes to the strategies do not affect the past performance, unless noted.
Groups DE1 and WE1: Daily and Weekly Trend Program for Stocks and ETFs
The Trend program seeks long-term directional changes in markets and the portfolios choose stocks and ETFs that have realized profitable performance over many years combined with good short-term returns.
Small gains for stocks, small losses for ETFs were still better than the benchmark SPY performance for December. For the year, the aggressive 10 stock portfolio finish up more than 20% and the larger 30 stock portfolio up 13%, both very much in line with the 12 year average returns. The ETF program, which attempts to select the strongest sectors, finished the year up over 5%. These returns are particularly good if you consider the defensive action necessary when the market sold off sharply in October. It is always easy to say in retrospect that the fast recovery means that trading programs should have been fully committed again, but that’s not realistic. We should remember that the index market recovered all its losses from the 2008 subprime crisis, but many individual investors reduced their positions during 2008 and never saw the recovery. Theory and reality often diverge.
The Weekly Trend Program performed better than the Daily Program in stocks, finishing the year up about 17% in both smaller and larger portfolios. The ETFs did not fare as well, both portfolios up for December but slightly down of the year. We need to point out that weekly trading signals have both advantages and disadvantages. During the October sell-off and rally it did better than the daily program by holding many of the positions. On the other hand, weekly signals will be late for a sell-off that continues. In the long run, equity markets are very noisy, and do not continue in one direction long without a correction. If you look further down at the sector performance for the Trend program using futures, you will see that the equity index sector is one of the few that did not post large profits in 2014. We attribute that to its underlying level of noise.
Group DE2: 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, often the way prices had moved before the period of uncertainty.
The Divergence program was profitable for the year, but did not reach its average returns. It does offer excellent diversification because it does not track the trend, so its benefit in a portfolio is greater than the returns. On its own, all three portfolios were positive for the year. One reason for lower performance is the lower exposure. It is not always possible to find enough divergence signals, even out of the large set of stocks that we track for this program. Therefore, the risk is also lower This is especially true for the ETF program, which draws from a much smaller set of about 70 ETFs. Still, the ETF returns have moved steadily higher on low risk.
Group DE3: 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 market direction for profits, although these portfolios are long-only because they are most often used in retirement accounts. When the broad market index turns down this program hedges part of the portfolio risk. The ETF Rotation program buys undervalued sectors, expecting them to outperform the other sectors over the short-term.
The Timing Program buys undervalued stocks so that it will buy the weakest even in a declining market until that stock shows that it is not expected to rally. Risk is protected with an absolute stop of 15% and also by hedging the broad index.
The Timing program is our one disappointment this year, returning losses in all portfolios. A downturn in the SPY generated a hedge position in December, which then dampened performance when the market failed to continue lower. At the very end of December we set one leg of a hedge in QQQ. While we hope that stocks rally and we take a small loss in this hedge, risk protection is an integral part of the Timing program. During a sustained downturn it will hold losses to a minimum compared to other program.
One of the issues with algorithmic trading is that it can be frustrating when it’s not making money, and requires discipline to consistently follow the trading signals. It is not possible to predict when the system will turn and returns will improve, and not possible to select when to trade and when not to trade. Case in point is the Trend Futures program this year. No one would have expected a 50% return after three years of sideways performance for the Industry.
We’ve been amazed at the consistency of the futures performance all year, and those returns continued into December, finishing the year 40% to 55% higher for the daily program and 42% to 60% higher for the weekly program. We admit that it’s not likely to continue, but we would have said that earlier in the year as well.
Looking at the sector returns for the Trend program below, only Equity Index shows a loss. That’s actually consistent with the long-term performance of equity futures. While they serve as excellent diversification and occasionally have a good positive run (remember the short positions of 2008?), they have too much noise for a trend program to identify turns in a timely fashion. Still, a small loss in equities during 2014 did little to dampen overall performance.
We can see that interest rates were profitable due to the Fed driving yields lower all year, FX due to a stronger dollar, energy because of a major drop in oil prices, metals because of the continued slide in gold (also due to the strength in the US dollar), and ags because of continued lower prices. Trends in (nearly) all sectors are the only way that the program can post such large returns.
The Divergence program was mixed for the year, with the larger portfolios doing well, but the smallest posting a net loss. In terms of sectors, the chart below shows that gains were limited to bonds and ags. Again we see the equity index markets posting the largest loss.
Groups DF1 and WF1: Daily and Weekly Trend Programs for Futures
Futures allow both high leverage and true diversification. The larger portfolios, such as $1million, 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 interested in lower leverage can simply scale all positions equally in proportion to their volatility preference. Note that these portfolios do not trade Asian futures, which we believe are more difficult for U.S. investors to execute.
Using the same strategy and portfolio logic, the Weekly Trend Program for Futures has the added smoothing resulting from looking only at Friday prices. While it will show a larger loss when the trend actually turns, most price moves are varying degrees of noise which this method can overlook.
Please read the new report describing our revised portfolio allocation methodology. It can be found in the drop-down menu under “Articles.”
We are not the only futures program doing well. Newedge shows that the CTA index rose more than 15% with trend followers up 19% for 2014. The BTOP 50 represents the top 50 CTAs, with an average of 12.4%.
The charts below show the daily and weekly Trend program performance. We recognize that the extreme move that we see for 2014 won’t be sustained, but we have no way of knowing when it will end. The advantage of an algorithmic system is that it will exit when the trend turns and will not make excuses for holding the positions “just in case” it rallies again.
Group DF2: Daily Divergence Portfolio for Futures
The Divergence program did well in the larger portfolio and maintains a nice upward trend in equity. It may be overlooked this year as the Trend program continues to post outstanding returns, but it is a sound program on its own. We’ll revisit this as the year takes shape.
Copyright 2015-Kaufman Signals January Performance Report-All Rights Reserved