Today’s article is the fifth installment of the Trend Composite strategy. The series was put on hold because of recent events and market volatility, but continues today and will complete before the end of the month. Today’s installment uses the highly curated All Weather ETF List, which covers most assets and has minimal overlap. We will show performance for different portfolios using Trend Composite cross signals and in-state signals. The latter insure exposure to bull markets as soon as the Market Regime changes and offer higher returns. We will also show how to use the ETF Trend Signal and Ranking Table to run this strategy.
Trend Composite Strategy Series
- Introduction to the Trend Composite and its Indicators (1-Feb)
- Trend Composite Signal and StochClose Ranking Table (2-Feb)
- All Signals Tests and Indicator Comparison (8-Feb)
- Trend Portfolio: TrendComp Crosses, StochClose Rank, Mkt Regime (15-Feb)
- Momentum Portfolio: Trend Status, StochClose Rank, Bear Mkt ETFs (22-Feb)
- The All Weather Portfolio, Bull-Bear Market ETFs, Market Regime (17-Mar)
- Adding an ATR Trailing Stop and Time Exit (24-Mar)
- Live Example when Market Regime Turns Bullish (28-Mar).
- Adding a Percent Based Profit Target (31-Mar)
Note that the Trend Composite is part of the TIP Indicator Edge Plugin for StockCharts ACP. I use Amibroker because this is where I can build the indicators, test the indicators and fully customize the chart views.
The charts are linked to a SharpChart with corresponding indicators or the SharpChart alternatives. Note that CCI is used instead of CCI Close, Full Stochastic (125,5,1) is used instead of StochClose (125,5) and the thick black line is the 125-day SMA.
A Review of the Prior Articles
The Trend Composite indicator generates trend signals when it turns positive or negative (new uptrend or new downtrend). The Trend Composite is covered in detail in part one.
StochClose is used to rank ETFs that have existing uptrends (positive Trend Composite). When faced with a choice, chartists can choose the ETF with the highest StochClose value. StochClose ranking is covered in part three.
The Composite Breadth Model defines the Market Regime: bull market for stocks or bear market for stocks. Bullish signals from stock-related ETFs are ignored when the Market Regime is bearish. The Market Regime is covered in part three.
The ETF universe can be divided into two parts: stock-related ETFs and non-stock ETFs. Non-stock ETFs are ETFs related to bonds, commodities and currencies. Bullish signals from stock-related ETFs are not taken when the Market Regime is bearish. Bullish signals from non-stock ETFs are taken because these ETFs are not correlated to the stock market.
There are two types of signals. First, buy and sell signals can be based on the Trend Composite turning positive or negative (new uptrend or new downtrend). Second, buy and sell signals can be based on the StochClose rank for ETFs with uptrends (positive Trend Composite). Buy the top ranked ETFs with uptrends, sell when the Trend Composite turns negative and replace with the top ranked ETF with an uptrend. This trend-momentum strategy is covered in part four and explained again in this article.
The All Weather List
It is important to define our ETF universe and make sure this universe aligns with our objectives. The first step is to narrow the universe. ETFDB.com shows 1903 passively managed non-leveraged ETFs. Of these, 1388 are equity-related ETFs, which is around 72% of the universe.
The Master ETF List at TrendInvestorPro has 276 ETFs. 227 (82%) are equity-related ETFs and 49 (18%) are non-equity ETFs. This list is clearly heavy on the equity side. Note that this list includes a few actively managed ETFs (ARKK et al) and some 55 non-US stock ETFs. I am not a fan of actively managed ETFs for systems because their managers often average down (add to their losers). This defeats the purpose of a trend-momentum strategy because such strategies do not add to losing positions when the trend turns down and the cheap get cheaper.
The All Weather List is a highly curated list of 50 ETFs with 36 stock-related ETFs (74%) and 14 non-stock ETFs (26%). This is a minimal list designed to cover most areas of the market and still allow for true diversification.
The vast majority of these ETFs have trading history back to at least 2006. HYG, UUP, FXY, IEI, DBE, DBB and DBA have data back to the first half of 2007. A long trading history is important when testing. Note that the majority of ETFs were launched after 2011.
Of the 36 stock-related ETFs in the All Weather List, there are 6 broad market ETFs (SPY, MDY, IJR, QQQ, RSP, IWC), 10 sector SPDRs, the REIT ETF (IYR), 19 industry group ETFs and one dividend ETF (DVY). The High-Yield Bond ETF (HYG) sometimes acts more like a stock-related ETF, but I decided to group it with the non-stock ETFs because it is not composed of stocks.
The 14 stock alternatives include six bond-related ETFs (AGG, IEI, IEF, TLT, LQD, HYG). There are five commodity-related ETFs (DBE, GLD, SLV, DBB, DBA) and three currency ETFs (UUP, FXE, FXY).
36 Stock ETFs:
SPY, RSP, MDY, IJR, IWC, QQQ, DVY, XLK, XLY, XLF, XLI, XLC, XLV, XLP, XLE, XLB, XLU, IYR, FDN, SOXX, IGV, ITB, XRT, KIE, KRE, ITA, PBW, IBB, IHF, IHI, XES, XOP, GDX, XME, PHO, PBJ
14 Alternatives:
AGG, IEI, IEF, TLT, LQD, HYG, GLD, SLV, DBE, DBB, DBA, FXE, FXY, UUP
Even though a few equity-related ETFs will buck the trend in a bear market, most are highly correlated to the broad market environment and picking winners in a bear market is very difficult. We do not gain much diversification benefit from owning a wide array of equity-related ETFs.
Also note that we do not really need several S&P 500 ETFs, five tech sector ETFs, three aerospace & defense ETFs, three semiconductor ETFs or ten clean energy ETFs. Sure, there will be some performance differences within these groups, but the correlations are extremely high at the group level. Again, there is NO diversification benefit from owning three different semiconductor ETFs.
Strategy Parameters
- Test period: 1/1/2007 to 1/31/2022 (15 years)
- All Weather ETF List for Universe
- 14 Equal-weight Positions
- Buy/Sell based on opening prices the next day
- Commission: $5 per trade (buy/sell = two trades)
- Slippage: 2.5 bps or .0025 percent per trade
- No dividends (article explaining why)
The performance tables shows the Compound Annual Return (CAR), Maximum Drawdown (MDD), Win Rate and some other metrics. The Gain/Loss Ratio is the Average Gain divided by the Average Loss. This one is pretty straight forward. Trend-followers would like to see this ratio above 3.
Profit Factor equals the number of winners multiplied by Average Gain divided by the number of losers multiplied by the Average Loss. A Profit Factor of 2 means total profits are outpacing total losses by a two to one margin.
Setting Benchmark Returns
The first table shows buy-and-hold SPY and all 50 All Weather ETFs. In other words, a portfolio of 50 equal weight positions in each of the All Weather ETFs. The Compound Annual Return (CAR) was a whopping 7.9% for SPY and 6.9% for the 50 ETF portfolio. Note, however, that the drawdowns were very large, 56.5% and 42%. Also note that 20% of the ETFs (10) in the All Weather list finished with a loss. Not everything goes up in the long-term.
Buy/Sell Trend Composite Crosses
The next test will buy and sell when the Trend Composite crosses above/below zero. StochClose is used as a tie-break when there are more signals than available positions. A buy entry occurs on the day after the Trend Composite crosses from negative to positive territory. A sell exit occurs the day after a cross into negative territory.
There are three tests. First, simple cross signals without a Market Regime filter. This means taking bullish signals in all 50 ETFs throughout the entire period. Second, a Market Regime filter is applied and bullish signals in all 50 ETFs are taken only when the Market Regime is bullish. Third, there is a Market Regime Filter and the 50 ETFs are divided into two groups (Bull Market ETFs and Bear Market ETFs). This means taking bullish signals in all ETFs when the Market Regime is bullish and taking bullish signals only in bear market ETFs when the Market Regime is bearish.
The first line shows the benchmark testing using Trend Composite crosses. The second test adds the Market Regime filter and the third test divides the ETFs into bull-bear groups (BB). The Compound Annual Return and Win Rates are similar. The lower drawdown is the main benefit. Notice how the Maximum Drawdown (MDD) went from -23.81% to 17.15% to 14.86%. Adding a Market Regime improved the MDD and dividing the ETFs into bull-bear groups improved the MDD even further.
Note that the equity-related ETFs are not sold indiscriminately when the Market Regime turns bearish. They are sold when/if their Trend Composite turns negative (sell signal).
Buy/Sell Trend Composite State and StochClose Rank
The next tests will use StochClose to rank uptrends and select ETFs. First, the ETF must be in an uptrend (Trend Composite positive). The strategy then selects the top ETFs based on their StochClose values. The Trend Composite insures there is an uptrend and StochClose rank insures that the ETF is a leader.
Again, the first test will start with no Market Regime, the second will add a Market Regime and the third will divide the ETFs into bull-bear groups.
Compared to the cross signals in the first table, the Compound Annual Returns are much higher and the Maximum Drawdowns are roughly similar. Note that there were 220 total trades when trading all 50 ETFs with a Market Regime filter. This is because there was no trading when the Market Regime was bearish. The total trades increase to 270 when the ETFs were divided and stock alternatives were traded during bear markets. This also led to more Exposure (invested 90% of the time).
Exposure also increased because the strategy tops up the portfolio as soon as the Market Regime turns bullish. For example, there were just nine long positions on May 29, 2020, the day the Market Regime turned bullish. This meant that stock-related ETFs became eligible and the strategy added the five top ranked stock ETFs on June 1st (FDN, IGV, XLV, QQQ, IBB).
In-state signals are important when the Market Regime moves from bearish to bullish. A strategy that waits for the Trend Composite to cross into positive territory would have missed the crosses in the ETFs above because their signals already triggered. Thus, the cross strategy would have had to wait for the next Trend Composite cross into positive territory. This means less invested at the outset of the new bull market.
Equity Curve, Drawdowns and Return Breakdown
The next chart shows the equity curve for the last strategy (green line) and buy-and-hold for SPY (blue line). This ETF strategy significantly outperformed buy-and-hold because it participated in bull markets and preserved capital during bear markets.
The blue shading on the next chart shows the drawdowns over the years. A drawdown is the percentage decline from an equity high to an equity low. There were some 12+ percent drawdowns from 2010 to 2012. The drawdowns were slightly less in the following years as the dips did not exceed 12% from 2013 to 2022.
The final table shows the annual and monthly return data. Of the 15 full years (2007 to 2021), there are four years with positive returns that were less than 5.3% and two years with negative returns (-1.4% and -3%). This means returns were subpar 40% of the time. The strategy made up for these subpar years with gains ranging from 7% to 35.6% in nine of the fifteen years.
Working with the Trend and Rank Table
The image below shows the Trend Signals and StochClose Ranking table (last close is 16 March). Here are the steps to find signals. First, enter “AllW” in the search box to isolate the All Weather ETFs. Second, hold the shift key, click “Trend” to sort with UPtrend at the top and then click “StochClose” to sort the UPtrends by the StochClose values. Hold the shift key from the first click until after the second click for this two column sort. This sorts the uptrends by StochClose and puts the leaders at the top.
The green outline on the symbols highlights the top 14 and half of these are equity related (XLE, XLU, XOP, XME, XES, DVY, ITA). Three of these are related to energy (XLE, XOP, XES). There are no bond-related ETFs because they are all in downtrends. Bonds are not an alternative to stocks when they are in downtrends. The absence of bonds is perhaps the reason XLU and DVY are in the top grouping.
Conclusions
Trading and investing is all about improving the odds and staying the course. Stocks are the best game in town during bull markets, but we need a plan for bear markets. A Market Regime filter can preserve capital when stocks are deemed too risky. A curated ETF list with stock-alternatives can give us access to non-correlated ETFs during bear markets. When the bear market ends, using in-state signals with a StochClose ranking filter can provide immediate exposure to stocks.
The hardest part of running a trend-momentum strategy is sticking with it through thick and thin. There will be periods of underperformance and it will seem like the strategy is broken, but these are often followed by periods of outperformance. Just when you least expect it. It is also normal to have periods of underperformance after periods of exceptional performance. Perhaps that is what we will seeing here in 2022 because the prior two years were exceptional.