Category: Trend Page 1 of 7

Is Managed Futures Value-able?

On June 19, 2023

In Craftsmanship, Inflation, Portfolio Construction, Trend, Value

In Return Stacking^TM: Strategies for Overcoming a Low Return Environment, we advocated for the addition of managed futures to traditionally allocated portfolios. We argued that managed futures’ low empirical correlation to both equities and bonds and its historically positive average returns makes it an attractive diversifier. More specifically, we recommended implementing managed futures as an overlay to a portfolio to avoid sacrificing exposure to core stocks and bonds.

The luxury of writing research is that we work in a “clean slate” environment. In the real world, however, investors and allocators must contemplate changes in the context of their existing portfolios. Investors rarely just hold pure beta exposure, and we must consider, therefore, not only how a managed futures overlay might interact with stocks and bonds, but also how it might interact with existing active tilts.

The most common portfolio tilt we see is towards value stocks (and, often, quality-screened value). With this in mind, we want to briefly explore whether stacking managed futures remains attractive in the presence of an existing value tilt.

Diversifying Value

If we are already allocated to value, one of our first concerns might be whether an allocation to managed futures actually provides a diversifying return stream. One of our primary arguments for including managed futures into a traditional stock/bond portfolio is its potential to hedge against inflationary pressures. However, there are arguments that value stocks do much of the same, acting as “low duration” stocks compared to their growth peers. For example, in 2022, the Russell 1000 Value outperformed the broader Russell 1000 by 1,145 basis points, offering a significant buoy during the throes of the largest bout of inflation volatility in recent history.

However, broader empirical evidence does not actually support the narrative that value hedges inflation (see, e.g., Baltussen, et al. (2022), Investing in Deflation, Inflation, and Stagflation Regimes) and we can see in Figure 1 that the long-term empirical correlations between managed futures and value is near-zero.

(Note that when we measure value in this piece, we will look at the returns of long-only value strategies minus the returns of broad equities to isolate the impact of the value tilt. As we recently wrote, a long-only value tilt can be effectively thought as long exposure to the market plus a portfolio that is long the over-weight positions and short the under-weight positions¹. By subtracting the market return from long-only value, we isolate the returns of the active bets the tilt is actually taking.)

Figure 1: Excess Return Correlation

Source: Kenneth French Data Library, BarclayHedge. Calculations by Newfound Research. Performance is backtested and hypothetical. Performance is gross of all costs (including, but not limited to, advisor fees, manager fees, taxes, and transaction costs) unless explicitly stated otherwise. Performance assumes the reinvestment of all dividends. Past performance is not indicative of future results. See Appendix A for index definitions.

Correlations, however, do not tell us about the tails. Therefore, we might also ask, “how have managed futures performed historically conditional upon value being in a drawdown?” As the past decade has shown, underperformance of value-oriented strategies relative to the broad market can make sticking to the strategy equally difficult.

Figure 2 shows the performance of the various value tilts as well as managed futures during periods when the value tilts realized a 10% or greater drawdown².

Figure 2: Value Relative Drawdowns Greater than 10%

We can see that while managed futures may not have explicitly hedged the drawdown in value, its performance remained largely independent and accretive to the portfolio as a whole.

To drive the point of independence home, we can calculate the univariate regression coefficients between value implementations and managed futures. We find that the relationship between the strategies is statistically insignificant in almost all cases. Figure 3 shows the results of such a regression.

Figure 3: Univariate Regression Coefficients

Source: Kenneth French Data Library, BarclayHedge. Calculations by Newfound Research. *, **, and *** indicate statistical significance at the 0.05, 0.01, and 0.001 level. Performance is backtested and hypothetical. Performance is gross of all costs (including, but not limited to, advisor fees, manager fees, taxes, and transaction costs) unless explicitly stated otherwise. Performance assumes the reinvestment of all dividends. Past performance is not indicative of future results. See Appendix A for index definitions.

But How Much?

As our previous figures demonstrate, managed futures has historically provided a positively diversifying benefit in relation to value; but how can we thoughtfully integrate an overlay into an portfolio that wants to retain an existing value tilt?

To find a robust solution to this question, we can employ simulation techniques. Specifically, we block bootstrap 100,000 ten-year simulated returns from three-month blocks to find the robust information ratios and MAR ratios (CAGR divided by maximum drawdown) of the value-tilt strategies when paired with managed futures.

Figure 4 shows the information ratio frontier of these portfolios, and Figure 5 shows the MAR ratio frontiers.

Figure 4: Information Ratio Frontier

Figure 5: MAR Ratio Frontier

Under both metrics it becomes clear that a 100% tilt to either value or managed futures is not prudent. In fact, the optimal mix, as measured by either the Information Ratio or MAR Ratio, appears to be consistently around the 40/60 mark. Figure 6 shows the blends of value and managed futures that maximizes both metrics.

Figure 6: Max Information and MAR Ratios

In Figure 7 we plot the backtest of a 40% value / 60% managed futures portfolio for the different value implementations.

Figure 7: 40/60 Portfolios of Long/Short Value and Managed Futures

These numbers suggest that an investor who currently tilts their equity exposure towards value may be better off by only tilting a portion of their equity towards value and introducing a managed futures overlay onto their portfolio. For example, if an investor has a 60% stock and 40% bond portfolio and the 60% stock exposure is currently all value, they might consider moving 36% of it into passive equity exposure and introducing a 36% managed futures overlay.

Depending on how averse a client is to tracking error, we can plot how the tracking error changes depending on the degree of portfolio tilt. Figure 8 shows the estimated tracking error when introducing varying allocations to the 40/60 value/managed futures overlay.

Figure 8: Relationship between Value/Managed Futures Tilt and Tracking Error

For example, if we wanted to implement a tilt to a quality value strategy, but wanted a maximum tracking error of 3%, the portfolio might add an approximate allocation of 46% to the 40/60 value/managed futures overlay. In other words, 18% of their equity should be put into quality-value stocks and a 28% overlay to managed futures should be introduced.

Using the same example of a 60% equity / 40% bond portfolio as before, the 3% tracking error portfolio would hold 42% in passive equities, 18% in quality-value, 40% in bonds, and 28% in a managed futures overlay.

What About Other Factors?

At this point, it should be of no surprise that these results extend to the other popular equity factors. Figures 8 and 9 show the efficient information ratio and MAR ratio frontiers when we view portfolios tilted towards the Profitability, Momentum, Size, and Investment factors.

Figure 9: Information Ratio Frontier for Profitability, Momentum, Size, and Investment Tilts

Figure 10: MAR Ratio Frontier for Profitability, Momentum, Size, and Investment Tilts

Figure 11: Max Information and MAR Ratios for Profitability, Momentum, Size, and Investment Tilts

Once again, a 40/60 split emerges as a surprisingly robust solution, suggesting that managed futures has historically offered a unique, diversifying return to all equity factors.

Conclusion

Our analysis highlights the considerations surrounding the use of managed futures as a complement to a traditional portfolio with a value tilt. While value investing remains justifiably popular in real-world portfolios, our findings indicate that managed futures can offer a diversifying return stream that complements such strategies. The potential for managed futures to act as a hedge against inflationary pressures, while also offering a diversifying exposure during relative value drawdowns, strengthens our advocacy for their inclusion through a return stacking^TM framework.

Our examination of the correlation between managed futures and value reveals a near-zero relationship, suggesting that managed futures can provide distinct benefits beyond those offered by a value-oriented approach alone. Moreover, our analysis demonstrates that a more conservative tilt to value, coupled with managed futures, may be a prudent choice for inverse to tracking error. This combination offers the potential to navigate unfavorable market environments and potentially holds more of a portfolio benefit than a singular focus on value.

Appendix A: Index Definitions

Book to Market – Equal-Weighted HiBM Returns for U.S. Equities (Kenneth French Data Library)

Profitability – Equal-Weighted HiOP Returns for U.S. Equities (Kenneth French Data Library)

Momentum – Equal-Weighted Hi PRIOR Returns for U.S. Equities (Kenneth French Data Library)

Size – Equal-Weighted SIZE Lo 30 Returns for U.S. Equities (Kenneth French Data Library)

Investment – Equal-Weighted INV Lo 30 Returns for U.S. Equities (Kenneth French Data Library)

Earnings Yield – Equal-Weighted E/P Hi 10 Returns for U.S. Equities (Kenneth French Data Library)

Cash Flow Yield – Equal-Weighted CF/P Hi 10 Returns for U.S. Equities (Kenneth French Data Library)

Dividend Yield – Equal-Weighted D/P Hi 10 Returns for U.S. Equities (Kenneth French Data Library)

Quality Value – Equal-Weighted blend of BIG HiBM HiOP, ME2 BM4 OP3, ME2 BM3 OP3, and ME2 BM3 OP4 Returns for U.S. Equities (Kenneth French Data Library)

Value Blend – An equal-weighted Returns of Book to Market, Earnings Yield, Cash Flow Yield, and Dividend Yield returns for U.S. Equities (Kenneth French Data Library)

Passive Equities (Market, Mkt) – U.S. total equity market return data from Kenneth French Library.

Managed Futures – BTOP50 Index (BarclayHedge). The BTOP50 Index seeks to replicate the overall composition of the managed futures industry with regard to trading style and overall market exposure. The BTOP50 employs a top-down approach in selecting its constituents. The largest investable trading advisor programs, as measured by assets under management, are selected for inclusion in the BTOP50. In each calendar year the selected trading advisors represent, in aggregate, no less than 50% of the investable assets of the Barclay CTA Universe.

What Is Managed Futures?

By Steven Braun

On February 2, 2023

In Craftsmanship, Portfolio Construction, Risk Management, Trend

Summary

Much like in 2008, managed futures as an investment strategy had an impressive year in 2022. With most traditional asset classes struggling to navigate the inflationary macroeconomic environment, managed futures has been drawing interest as a potential diversifier.
Managed futures is a hedge fund category that uses futures contracts as their primary investment vehicle. Managed futures managers can engage in many different investment strategies, but trend following is the most common.
Trend following as an investment strategy has a substantial amount of empirical evidence promoting its efficacy as an investment strategy. There also exist several behavioral arguments for why this anomaly exists, and why we might expect it to continue.
As a diversifier, multi-asset trend following has provided diversification benefits when compared to both stocks and bonds. Additionally, trend following has posted positive returns in the four major drawdowns in equities since 2000.

Cut short your losses, and let your winners run. – David Ricardo, 1838

What is Managed Futures?

Managed futures is a hedge fund category originating in the 1980s, named for the ability to trade (both long and short) global equity, bond, commodity, and currency futures contracts. Today, these strategies have been made available to investors in both mutual fund and ETF wrappers. The predominate strategy of most managed futures managers is trend following, so much so, that the terms are often used synonymously.

While trend following is by far the largest and most pronounced strategy in the category, it is not the only strategy used in the space.¹ Managed futures can engage in trend following, momentum trading, mean reversion, carry-focused strategies, relative value trading, macro driven strategies, or any combination thereof. Any individual managed futures manager may have a certain bias towards one of the strategies, though, trend following is by far the most utilized strategy of the group².

Figure 1: The Taxonomy of Managed Futures

Adapted from Kaminski (2014). The most common characteristics are highlighted in orange.

What is Trend Following?

Simply put, trend following is a strategy that buys (‘goes long’) assets that have been rising in price and sells (‘goes short’) assets that have been decreasing in price, based on the premise that this trend will continue. The precise method of measuring trends varies widely, but each primarily relies on the difference between an asset’s price today and the price of the same asset previously. Some common methods of measuring trends include total return measurements, moving averages, and regression lines. These different approaches are all mathematically linked, and empirical evidence does not suggest that one method is necessarily better than another³.

Trend following has a rich history in financial markets, with centuries of evidence supporting the idea that markets tend to trend. The obvious question to then ask is: why? The past few decades of academic research has focused on explaining theories such as the Efficient Market Hypothesis and research into explanatory market factors (such as value and size), diminishing the amount of research being conducted on trend following.

Figure 2: The Life Cycle of a Trend

Adapted from AQR. For illustrative purposes only.

The classification of trend following as an anomaly, however, has not left it without theories for why it works. There are a number of generally accepted explanations for why trend following works, and more importantly, why the anomaly might continue to persist.

Anchoring Bias: When new data enters the marketplace, investors can overly rely on historical data, thereby underreacting to the new information. This can be seen in Figure 3 where, after the catalyst of new information enters the market, the price of a security will directionally follow the fair value of the asset, but not with a large enough magnitude to match the fair value precisely.

Disposition Effect: Investors have a tendency to take gains on their winning positions too early and hold onto their losing positions too long.

Herding: After a noticeable trend has been established, investors “bandwagon” into the trade, prolonging the directional trend, and potentially pushing the price past the asset’s fair value⁴.

Confirmation Bias: Investors tend to ignore information that is contrary to an their beliefs. A positive (or negative) signal will be ignored if the investor has a differing view, extending the time frame for the convergence of an asset’s price to its fair value.

Rational Inattention Bias: Investors cannot immediately digest all information due to a lack of information processing resources (or mental capacity). Consequently, prices move towards fair value more slowly as the information is processed by all investors.

As previously mentioned, methodologies may vary widely when analyzing an asset’s trend, but the general theme is to view an asset’s current price relative to some measure of its recent history. For example, one common example of this is to observe an asset’s current price versus its 200-day moving average: initiating a long position when the price is above its moving average or a short position when it is below. Extending Figure 2, we can graphically depict the trade cycle attempting to take advantage of such a trend.

Figure 3: The Life Cycle of a Trade

Source: Newfound Research, AQR. For illustrative purposes only

Of course, using such an idealized description of a trend is not typically what is found in the market, which leads to many false-starts, The risk-management decisions made to reduce the impact of these false-starts begins to highlight part of the attractiveness of the strategy as a diversifier.

Consider that the fair value of an asset is generally never known with a high degree of certainty. A trend following manager is thus reliant on the perceived direction of trend at any given time, and so, must make choices based on how the trend evolves or not.

Figure 4: Heads I Trend, Tails I Don’t

Adapted from Michael Covel. For illustrative purposes only.

When the model indicates that a trend has formed, the manager will initiate a position in the direction of the indicated trend (either short or long – blue line in Figure 4). As long as the trend continues, the strategy will hold that position, and only exit when the signal indicates that the trend no longer exists. At that time, the manager will remove the position, potentially taking the opposite position⁵.

The second case (red line in Figure 4) is one in which the trend reverses shortly after a position has been initiated. After establishing a position in the asset, the price of the asset reverts to its previous levels, possibly completely reversing in direction. In such a case, the signal will indicate that the trend no longer exists and recommend that the position be removed.

Historically, by quickly cutting losers and letting winning trades run, trend following has created a positively skewed return profile. Managed futures strategies tend to trade many different markets and underlying assets. This minimizes the impact of trends being rejected but may increase the probability of taking a position in an asset that has an outlier trend occurring that might be out of the scope of a traditional portfolio.

Kaminski (2014) refers to this characteristic as divergent risk taking⁶, where a divergent investor “profess[es] their own ignorance to the true structure of potential risks/benefits with some level of skepticism for what is knowable or is not dependable”.

This divergent risk behavior results in a positively skewed return distribution by not risking too much on a trade, removing the position if it goes against you, and allowing a trade to run if it is winning⁷.

The structural nature of trend following minimizes the size of any bets taken, and quickly eliminates a position if the bet is not paying off. By diversifying across many markets, asset classes, and economic goods, while maintaining sensible positions without directional bias, the strategy maintains staying power by not swinging for the fences and staying with a time-proven approach⁸, in a well-diversified manner.

Using Managed Futures as A Diversifier

The traditional investor portfolio has typically been dominated by two assets: stocks and bonds. In recent history, investors have even been able to use fixed income to buffer equity risk as high-quality bonds have exhibited flight-to-safety characteristics in times of extreme market turmoil. In the first two decades of the 2000s, this pairing has worked extremely well given that interest rates declined over the period, inflation remained low, and the bonds were resilient during the fallout of the tech bubble and the Great Financial Crisis.

In Figure 5, we chart the relationship between the year-over-year Consumer Price Index for All Urban Consumers (“CPIAUCSL”) versus the 12-month correlation between U.S. Stocks and 10-Year U.S. Treasuries⁹. We can see that negative correlation is most pronounced when inflation is low. Positive correlation regimes, on the other hand, have historically occurred in all realized ranges of CPI changes, the most striking occurring when inflation was extraordinarily high.

Figure 5: The Relationship Between Inflation and Equity-Bond Correlation

Source: FRED, Kenneth French Data Library, Tiingo. For illustrative purposes only.

Since trend following can hold both long and short positions, it has the potential to trade price trends in assets in any direction that may emerge from increasing inflation risks. This is highlighted by the performance of trend following in 2022, where the year-to-date real returns of U.S. equities¹⁰, 10-Year U.S. Treasuries, and the SG CTA Trend Index as of December 31, 2022 , were -19.5%, -16.5%, and +27.4%, respectively. During 2022, trend following strategies were generally long the U.S. Dollar, short fixed income securities, and short equity indices. Additionally, the managers tended to hold mixed positions in the commodity space, taking long and short positions in the individual commodity contracts exhibiting both positive and negative trends.

Importantly, the dynamics exhibited throughout different economic regimes (such as monetary inflation vs supply/demand inflation) will unfold differently, so positions that were profitable in 2022 will likely not be the same in all environments. Trend following as a strategy, is dynamic in nature, and will adjust positioning as trends emerge and fade, regardless of the economic regime.

In addition to historically providing a ballast in inflationary regimes, one of managed futures’ claims to fame stems from the strategy’s ability to provide negative correlation in times of financial stress, specifically, in equity crises. The net result of including an allocation to trend following strategies during these periods has been a reduction in portfolio drawdowns and portfolio volatility.

Though managed futures have been in existence since the 1980’s, the strategy garnered its popularity coming out of the Great Financial Crisis, as it was one of the few investment strategies to provide a positive return. While this event shot the strategy to prominence, it was not an isolated incident. In fact, this relationship has been repeated frequently throughout history.

Table 1 shows the cumulative nominal returns of stocks, bonds, and managed futures when the equity market realized a greater-than 20% drawdown.

Table 1: Nominal Return of Equities, Bonds, and Managed Futures During Equity Crises

Source: FRED, Kenneth French Data Library, BarclayHedge. Calculations by Newfound Research. Time period is based on data availability. Performance is gross of all costs (including, but not limited to, advisor fees, manager fees, taxes, and transaction costs) unless explicitly stated otherwise Past performance is not a reliable indicator of future performance.

Since the inception of the SG CTA Trend Index¹¹, bonds have provided diversification benefits in three of the four large drawdowns. 2022, however, was the first period in which inflation has been a concern in the market, and U.S. Treasuries were insufficient to reduce risk in a traditional portfolio.

We can see, though, that the SG CTA Trend Index provided similar diversification benefits during the drawdowns in the first two decades of the century, but also proved capable while inflation shocks rose to prominence in 2022.

Figure 6: Performance From 1999 to 2022

Source: BarclayHedge, Tiingo. 60/40 Portfolio is the Vanguard Balance Index Fund (“VBINX”) and returns presented are net of the management fee of the fund. Time period is based on data availability. Performance is gross of all costs (including, but not limited to, advisor fees, manager fees, taxes, and transaction costs) unless explicitly stated otherwise. Past performance is not a reliable indicator of future performance.

Conclusion

Traditional portfolios consisting of equity and fixed income exposure have enjoyed two decades of strong performance due to favorable economic tailwinds. With the changing economic regime and uncertainty facing markets ahead, however, investors have begun searching for potential additions to their portfolios to protect against inflation and to provide diversifying exposure to other macroeconomic headwinds.

Trend following as a strategy has extensive empirical evidence supporting both its standalone performance, as well as the diversifying benefits in relation to traditional asset classes such as stocks and bonds. In addition, trend following is mechanically convex in that it can provide positive returns in both bull and bear markets.

Managed futures is a strong contender as an addition to a stock-and-bond heavy portfolio. Finding its roots in the 1980s, the strategy has a tenured history in the investment landscape with a demonstrated history of providing diversifying exposure in times of equity crisis.

In this paper, we have shown that trend following is a robust trading strategy with behavioral underpinnings, suggesting that the strategy has staying power in the long-run, as well as desirable characteristics due to the mechanical nature of the strategy.

As a potential addition to a traditional investment portfolio, managed futures provides a source of diversification beyond that of mainstream asset classes, as well as strong absolute returns on a standalone basis.

APPENDIX A: TREND FOLLOWING AS AN OPTIONS STRADDLE

A trend following strategy can benefit from both positive and negative price trends. If prices are increasing, then a long position can be initiated; if prices are decreasing, then a short position can be initiated. Said differently: a trend following strategy can potentially profit from both increases or decreases in price.

This characteristic is immediately reminiscent of a long position in an option straddle, where a put and call option are purchased with the same strike price. This option position would, thereby, benefit if the price moves largely either positive or negative¹².

Figure A1: Long Straddle Payoff Profile

Source: Newfound Research. For illustrative purposes only.

Empirically, these strategies have in fact performed remarkably similar. To illustrate this, we will create two simple strategies.

The first strategy is a simple trend following strategy that takes a long position in the S&P 500 when its prior 12-month return is positive, and a short position when its negative.

The second strategy will attempt to replicate the delta-position of a straddle expiring in one month, struck at the close price of the S&P 500 twelve months ago. We then compute the delta of this position using the Black-Scholes model¹³ and take a position in the S&P 500 equal to the computed delta. For example, if the price of the S&P 500 12-months ago was $3,000, we would calculate the delta of a straddle struck at $3,000. Since the delta of this position will range between -1 and 1, the strategy will use this as an allocation to the S&P 500.

Figure A2: Replicating Trend Following with Straddles

Source: Tiingo. Calculations by Newfound Research. Returns assume the reinvestment of all dividends. The S&P 500 is represented by the Vanguard 500 Index Fund Investor Shares (“VFINX”). For illustrative purposes only. Past performance is not a reliable indicator of future performance.

For both strategies, we will assume that any excess capital is held in cash, returning 0%. Figure A2 plots the growth of $1 invested in each strategy.

As we can see, the option strategy and the trend following strategy provide a roughly equivalent return profile. In fact, if we compare the quarterly returns of the two strategies to the S&P 500, an important pattern emerges. Both strategies exhibit convex relationships in relation to the S&P 500.

Figure A3: Trend Following Relationship to the Underlying

Source: Newfound Research. For illustrative purposes only.

Figure A4: Straddle Replication Relationship to the Underlying

Source: Newfound Research. For illustrative purposes only.

APPENDIX B: Index Definitions

U.S. Stocks: U.S. total equity market return data from Kenneth French Library. Performance is gross of all costs (including, but not limited to, advisor fees, manager fees, taxes, and transaction costs) unless explicitly stated otherwise. Performance assumes the reinvestment of all dividends.

10-Year U.S. Treasuries: The 10-Year U.S. Treasury index is a constant maturity index calculated by assuming that a 10-year bond is purchased at the beginning of every month and sold at the end of that month to purchase a new bond at par at the beginning of the next month. You cannot invest directly in an index, and unmanaged index returns do not reflect any fees, expenses, or sales charges. The referenced index is shown for general market comparison and is not meant to represent any Newfound index or strategy. Data for 10-Year U.S. Treasury yields come from the Federal Reserve of St. Louis economic database (“FRED”).

SG Trend Index: The SG Trend Index is designed to track the largest 10 (by AUM) CTAs and be representative of the managed futures trend-following space.

Option-Based Trend Following

By Nathan Faber

On June 23, 2020

In Risk Management, Trend, Weekly Commentary

This post is available as a PDF download here.

Summary

The convex payoff profile of trend following strategies naturally lends itself to comparative analysis with option strategies.
To isolate the two extremes of paying for whipsaw – either up front or in arrears – we replicate an option strategy that buys 1-month at-the-money calls and puts based on the trend signal.
We find that while option premiums steadily eat away at the balance of the options portfolio, the avoidance of large whipsaw events gives the strategy a boost at key times over the past 15 years, especially recently.
We examine how this whipsaw cost fits into the historical context of the options strategy and explore some simple ways to shift between the option-based trend following and the standard model.
The extent that whipsaw can be mitigated while still maintaining the potential to earn diversified returns is likely limited, but the optimal blend of trend following and options can be a beneficial guideline for investors to weather both sudden and prolonged drawdowns.

The non-linear payoff of trend following strategies has many similarities to options strategies, and by way of analogy, we can often gain insight into which market environments will favor trend following and why.

In our previous research piece, Straddles and Trend Following, we looked at purchasing straddles – that is, a call option and a put option – with a strike price tied to the anchor price of the trend following model. For example, if the trend following model invested in equities when the return over the past 12 months was positive, for a security that was at $100 12-months ago and is at $120 today, we would purchase a call and a put option with a strike price of $100. In this case, the call would be 20% in-the-money (ITM) and the put would be out-of-the-money (OTM).

In essence, this strategy acted like an insurance policy where the payout was tied to a reversion in the trend signal, and the premium paid when the trend signal was strong was small.

This concept of insurance is an important discussion topic in trend following strategies. The risk we must manage in these types of strategies, either directly through insurance or some other indirect means like diversification, is whipsaw.

In this commentary, we will construct an options strategy that is similar to a trend following strategy. The option strategy will pay a premium up-front to avoid whipsaw. By comparing this strategy to trend following that bears the full risk of whipsaw, we can set a better practical bound for how much investors should expect to pay or earn for bearing this risk.

Methodology and Data

For this analysis, we will use the S&P 500 index for equity returns, the 1-year LIBOR rate as the risk-free rate, and options data on the S&P 500 (SPX options).

To bridge the gap between practice and abstraction, we will utilize a volatility surface calibrated to real option data to price options. We will constrain our SPX options to $5 increments and interpolate total implied variance to get prices for options that were either illiquid or not included in the data set.

For the most part, we will stick to options that expire on the third Friday of each month and will mention when we deviate from that assumption.

The long/short trend equity strategy looks at total returns of equities over 12 months. If this return is positive, the strategy invests in equities for the following month. If the return is negative, the strategy shorts equities for the following month and earns the risk-free rate on the cash. The strategy is rebalanced monthly on the options expiration dates.

For the option-based trend strategy, on each rebalance date, we will purchase a 1-month call if the trend signal is positive or a put if the trend signal is negative. We will purchase all options at-the-money (ATM) and hold them to expiration. The strategy is fully cash-collateralized. Any premium is paid on the options roll date, interest is earned on the remaining account balance, and the option payout is realized on the next roll date.

Why are we now using ATM options when previous research used ITM and OTM options, potentially deeply ITM or OTM?

Here we are looking to isolate the cost of whipsaw in the premium paid for the option while earning a payout that is close to that of the underlying in the event that our trend signal is correct. If we utilized OTM options, then our premium would be lower but we would realize smaller gains if the underlying followed the trend. ITM options would have downside exposure before the protection kicked in.

We are also not using straddles since we do not want to pay extra premium for the chance to profit off a whipsaw. The underlying assumption here is that there is value in the trend following signal. Either strategy is able to capitalize on that (i.e. it’s the control variable); the strategies primarily differ in their treatment of whipsaw costs.

The High Cost of ATM Options

The built-in whipsaw protection in the options does not come cheap. The chart below shows the –L/S trend following strategy–, the –option-based trend strategy–, and the ratio of the two (dotted).Source: DiscountOptionsData.com. Calculations by Newfound Research. Returns are hypothetical and backtested. Returns are gross of all fees including, but not limited to, management fees, transaction fees, and taxes. Returns assume the reinvestment of all distributions.

During normal market environments and even in prolonged equity-market drawdown periods like 2008, trend following outperformed the option-based strategy. Earning the full return on the underlying equity is generally beneficial.

However, something that is “generally beneficial” can be erased very quickly. In March 2020, the trend following strategy reverted back to the level of the option-based strategy. If you had only looked at cumulative returns over those 15 years, you would not be able to tell much difference between the two.

The following chart highlights these tail effects.

Source: DiscountOptionsData.com. Calculations by Newfound Research. Returns are hypothetical and backtested. Returns are gross of all fees including, but not limited to, management fees, transaction fees, and taxes. Returns assume the reinvestment of all distributions.

In most months, the option-based strategy forfeits its ~1.5% premium for the ATM option. The 75^th percentile cutoff is 2.2% and the 90^th percentile cutoff is 2.9%. These premiums have occasionally spiked to 6-7%.

While these premiums are not always forfeited without some offsetting gain, they are always paid relative to the trend following strategy.

A 3% whipsaw event in trend should definitely not be a surprise based on the typical up-front cost of the option strategy.

Source: DiscountOptionsData.com. Calculations by Newfound Research.

But What About a 30% Whipsaw?

Now that’s a good question.

Up until March 2020, for the 15 years prior, the largest whipsaws relative to the options strategy were 12-13%. This is the epitome of tail risk, and it can be disheartening to think that now that we have seen 30% underperformance, we should probably expect more at some point in the (hopefully very distant) future.

However, a richer sample set can shed some light on this very poor performance.

Let’s relax our assumption that we roll the options and rebalance the trend strategies on the third Friday of the month and instead allow rebalances and rolls on any day in the month. Since we are dealing with one-month options, this is not beyond implementation since there are typically options listed that expire on Monday, Wednesday, and Friday.

The chart below shows all of these option strategies and how large of an effect that roll / rebalance timing luck can have.

With timing luck in both the options strategies and trend following, there can be large effects when the luck cuts opposite ways.

The worst returns between rebalances of trend following relative to each options strategy highlight how bad the realized path in March 2020 truly was.

In many of the trend following and option strategies pairs, the worst underperformance of trend following over any monthlong period was around 10%.

Returning to the premise that the options strategies are analogous to trend following, we see the same effects of timing luck that we have explored in previous research: effects that make comparing variants of the same strategy or similar strategies more nuanced. Whether an option strategy is used for research, benchmarking, or active investing, the implications of this timing luck should be taken into account.

But even without taking a multi-model approach at this point to the options strategy, can we move toward a deeper understanding of when it may be an effective way to offset some of the risk of whipsaw?

I’d Gladly Pay You Tuesday for a Whipsaw Risk Today

With the two extremes of paying for whipsaw up front with options and being fully exposed to whipsaw through trend following, perhaps there is a way to tailor this whipsaw risk profile. If the risk of whipsaw is elevated but the cost of paying for the insurance is cheap, then the options strategy may be favorable. On the other hand, if option premiums are high, trend following may more efficiently capture the market returns.

The price of the options (or their implied volatilities) is a natural place to start investigating this topic since it encapsulates the premium for whipsaw insurance. The problem is that it may not be a reliable signal if there is no barrier to efficiency in the options market, either behavioral or structural.

Comparing the ATM option implied volatilities with the trend signal (12-month trailing returns), we see a negative correlation, which indicates that the options-based strategy will have a higher hurdle rate of return in strongly downtrending market environments.

Source: DiscountOptionsData.com. Calculations by Newfound Research.

But this is only one piece of the puzzle.

Do these implied volatilities relate to the forward 1-month returns for the S&P 500?

Based on the above scatterplot: not really. However, since we are merely sticking implied volatility in the middle of the trend following signal and the forward return, and we believe that trend following works over the long run, then we must believe there is some relationship between implied volatility and forward returns.

While this monthly trend following signal is directionally correct over the next month 60% of the time, historically, that says nothing about the magnitude of the returns based on the signal.

Without looking too much into the data to avoid overfitting a model, we will set a simple cutoff of 20% implied volatility. If options cost more than that, we will utilize trend following. If they cost less, we will invest in the options strategy.

We will also compare it to a 50/50 blend of the two.

The switching strategy (gray line) worked well until around 2013 when the option prices were cheap, but the risk of whipsaw was not realized. It did make it through 2015, 2016 and 4Q 2018 better than trend following.

When viewed in a broader context of a portfolio, since these are alternative strategies, it does not take a huge allocation to make a difference. These strategies manage equity risk, so we can pair them with an allocation to the S&P 500 (SPY) and see how the aggregate statistics are affected over the period from 2005 to April 2020.

The chart below plots the efficient frontiers of allocations to 100% SPY at the point of convergence on the right of the graph) to 40% SPY on the left of the graph with the remainder allocated to the risk- management strategy.

The Sharpe ratio is maximized at a 35% allocation to the switching strategy, a 25% allocation to the option-based strategy, and 10% for the trend following strategy.

Conclusion

In this research note, we explored the link between trend following and options strategies using 1-month ATM put and call options, depending on the sign of the trend.

The cost of ATM options Is generally 1.5% of the portfolio value, but the fact that this cost can spike upwards of 9% should justify larger whipsaws in trend following strategies. Very large whipsaws, like in March 2020, not only show that the cost can be seemingly unbounded but also that there is significant exposure to timing luck based upon the option roll dates.

Then, we moved on to investigating a simple way to allocate between the two strategies based upon the cost of the options, When the options were cheap, we used that strategy, and when they were expensive, we invested in the trend following strategy. A modest allocation is enough to make a different in the realized efficient frontier.

Deciding to pay the up-front payment of the whipsaw insurance premium, bear the full risk a whipsaw, or land somewhere in between is largely up to investor preferences. It is risky to have a large downside potential, but the added benefit of no premiums can be enough to offset the risk.

An implied volatility threshold was a rather crude signal for assessing the risk of whipsaw and the price of insuring against it. Further research into one or multiple signals and a robust process for aggregating them into an investment decision is needed to make more definitive statements on when trend following is better than options or vice versa. The extent that whipsaw can be mitigated while still maintaining the potential to earn diversified returns is likely limited, but the optimal blend of trend following and options can be a beneficial guideline for investors to weather both sudden and prolonged drawdowns.

Straddles and Trend Following

By Nathan Faber

On May 11, 2020

In Momentum, Portfolio Construction, Risk Management, Trend, Weekly Commentary

This post is available as a PDF download here.

Summary

The convex payoff profile of trend following strategies naturally lends itself to comparative analysis with option strategies. Unlike options, however, the payout of trend following is not guaranteed.
To compare and contrast the two approaches, we replicate simple trend following strategies with corresponding option straddle strategies.
While trend-following has no explicit up-front cost, it also bears the full brunt of any price reversals. The straddle-based approach, on the other hand, pays an explicit cost to insure against sudden and large reversals.
This transformation of whipsaw risk into an up-front option premium can be costly during strongly trending market environments where the option buyer would have been rewarded more for setting a higher deductible for their implicit insurance policy and paying a lower premium.
From 2005-2020, avoiding this upfront premium was beneficial. The sudden loss of equity markets in March 2020, however, allowed straddle-based approaches to make up for 15-years of relative underperformance in a single month.
Whether an investor wishes to avoid these up-front costs or pay them is ultimately a function of the risks they are willing to bear. As we like to say, “risk cannot be destroyed, only transformed.”

We often repeat the mantra that, “risk cannot be destroyed, only transformed.” While not being able to destroy risk seems like a limitation, the assertion that risk can be transformed is nearly limitless.

With a wide variety of investment options, investors have the ability to mold, shape, skew, and shift their risks to fit their preferences and investing requirements (e.g. cash flows, liquidity, growth, etc.).

The payoff profile of a strategy is a key way in which this transformation of risk manifests, and the profile of trend following is one example that we have written much on historically. The convex payoff of many long/short trend following strategies is evident from the historical payoff diagram.

Source: Newfound Research. Payoff Diversification (February 10^th, 2020). Source: Kenneth French Data Library; Federal Reserve Bank of St. Louis. Calculations by Newfound Research. Returns are hypothetical and assume the reinvestment of all distributions. Returns are gross of all fees, including, but not limited to, management fees, transaction fees, and taxes. The 60/40 portfolio is comprised of a 60% allocation to broad U.S. equities and a 40% allocation to a constant maturity 10-Year U.S. Treasury index. The momentum portfolio is rebalanced monthly and selects the asset with the highest prior 12-month returns whereas the buy-and-hold variation is allowed to drift over the 1-year period. The 10-Year U.S. Treasuries index is a constant maturity index calculated by assuming a 10-year bond is purchased at the beginning of every month and sold at the end of that month to purchase a new bond at par at the beginning of the next month. You cannot invest directly in an index and unmanaged index returns do not reflect any fees, expenses or sales charges. Past performance is not indicative of future results.

This characteristic “V” shape in the diagram is reminiscent of an option straddle, where an investor buys a put and call option of the same maturity struck at the same price. This position allows the investor to profit if the price of the underlying security moves significantly in either direction, but they pay for this opportunity in the option premiums.

Source: theoptionsguide.com

The similarity of these payoff profiles is no coincidence. As we demonstrated in Trend – Convexity and Premium (February 11^th, 2019), simple total return trend following signals coarsely approximate the delta of the straddle. For those less familiar with the parlance of options, delta is the sensitivity in the value of the options to changes in the underlying stock. For example, if delta is +1, then the value of the option position will match price changes in the underlying dollar-for-dollar. If delta is -1, then the position will lose $1 for every dollar gained in the underlying and vice versa (i.e. the position is effectively short).

How does this connection arise? Consider a naïve S&P 500 trend strategy that rebalances monthly and uses 12-month total returns as a trend signal, buying when prior returns are positive and shorting when prior returns are negative. The key components of this strategy are today’s S&P 500 level and the level 12 months ago.

Now consider a strategy that buys a 1-month straddle with a strike equal to the level of the S&P 500 12 months ago. When the current level is above the strike, the strategy’s delta will be positive and when the level is below the strike, the delta will be negative. What we can see is that the sensitivity of our options trade to changes in the S&P 500 will match the sign of the trend strategy!

There are two key differences, however. First, our trend strategy was designed to always be 100% long or 100% short, whereas the straddle’s sensitivity can vary between -100% and 100%. Second, the trend strategy cannot change its exposure intramonth whereas the straddle will. In fact, if price starts above the strike price (a positive trend) but ultimately ends below – so far as it is sufficiently far that we can make up for the premium paid for our options – the straddle can still profit!

In this commentary, we will compare and contrast the trend and option-based approaches for a variety of lookback horizons.

Methodology and Data

For this analysis, we will use the S&P 500 index for equity returns, the iShares Short-term U.S. Treasury Bond ETF (ticker: SHV) as the risk-free rate, and monthly options data on the S&P 500 (SPX options).

The long/short trend equity strategy looks at total returns of equities over a given number of months. If this return is positive, the strategy invests in equities for the following month. If the return is negative, the strategy shorts equities for the following month and earns the short-term Treasury rate on the cash. The strategy is rebalanced monthly on the third Friday of each month to coincide with the options expiration dates.

For the (semi-equivalent) straddle replication, at the end of each month we purchase a call option and a put option struck at the level of the S&P 500 at the beginning of the lookback window of the trend following strategy. We can also back out the strike price using the current trend signal value and S&P 500. For example, if the trend signal is 25% and the S&P 500 is trading at $3000, we would set the strike of the options at $2400.

The options account is assumed to be fully cash collateralized. Any premium is paid on the options roll date, interest is earned on the remaining account balance, and the option payout is realized on the next roll date.

To value the options, we employ Black-Scholes pricing on an implied volatility surface derived from available out-of-the money options. Specifically, on a given day we fit a parabola to the implied variances versus log-moneyness (i.e. log(strike/price)) of the options for each time to maturity.

In prior research, we created straddle-derived trend-following models by purchasing S&P 500 exposure in proportion to the delta of the strategy. To calculate delta, we had previously priced the options using 21-day realized volatility as a proxy for implied volatility. This generally leads to over-pricing the options during crisis times and underpricing during more tame market environments, especially for deeper out of the money puts. In this commentary we are actually purchasing the straddles and holding them for one month.

Source: Tiingo and DiscountOptionData.com. Calculations by Newfound Research.

Straddle vs. Trend Following

Below we plot the ratio of the equity curves for the straddle strategies versus their corresponding trend following strategies. When the line is increasing, the straddle strategy is out-performing, and when the line is decreasing the trend strategy is out-performing.

Source: Tiingo and DiscountOptionData.com. Calculations by Newfound Research.

We can see, generally, that trend following out-performed the explicit purchase of options for almost all lookback periods for the majority of the 15-year test period.

It is only with the most recent expiration – March 20, 2020 – that many of the straddle strategies came to out-perform their respective trend strategies. With the straddle strategy, we pay an explicit premium to help insure our position against sudden and large intra-month price reversals. This did not occur very frequently during the 15 year history, but was very valuable protection in March when the trend strategies were largely still long coming off markets hitting all-time-highs in late February.

Shorter-term lookbacks fared particularly well during that month, as the trend following strategy was in a long position on the February 2020 options expiry date, and the straddles set by the short-term lookback window were relatively cheap from a historical perspective.

Source: Tiingo and DiscountOptionData.com. Calculations by Newfound Research.

Note the curious case of the 14-month lookback. Entering March, the S&P 500 was +45% over a 14-month lookback (almost perfectly anchored to December 2018 lows). Therefore, the straddle was struck so deep in the money that it did not create any protection against the market’s sudden and large drawdown.

Prior to March 2020, only the 8- and 15-month lookback window strategies had outperformed their corresponding trend following strategies. In both cases, it was just barely and just recently.

Another interesting point to note is that longer-term straddle strategies (lookbacks greater than 9 months) shared similar movements during many periods while shorter-term lookbacks (3-6 months) showed more dispersion over time.

Overall, many of the straddles exhibit more “crisis alpha” than their trend following counterparts. This is an explicit risk we pay to hedge with the straddle approach and a fact we will discuss in more detail later on.

How Equity Movements Affect Straddles

Before we move into a discussion of how we can frame the straddle strategies, it will be helpful to revisit how straddles are affected by changing equity prices and how this effect changes with different lookback windows for the strategies.

Consider the delta of a straddle versus how far away price is from the strike (normalized by volatility).

Naively, we might consider that the longer our trend lookback window – and therefore the further back in time we set our strike price – the further away from the strike that price has had the opportunity to move. Consider two extremes: a strike set equal to the price of the S&P 500 10 years ago versus one set a day ago. We would expect that today’s price is much closer to that from a day ago than 10 years ago.

Therefore, for a longer lookback horizon we might expect that there is a greater chance that the straddle is currently deeper in the money, leading to a delta closer to +/- 1. In the case of straddles struck at index levels more recently realized, it is more likely that price is close to at the money, leading to deltas closer to 0.

This also means that while the trend following strategy is taking a binary bet, the straddle is able to modulate exposure to equity moves when the trend is less pronounced. For example, if a 12-month trend signal is +1%, the trend model will retain a +1 exposure while the delta of the straddle may be closer to 0.

Source: Tiingo and DiscountOptionData.com. Calculations by Newfound Research.

Additionally, when the delta of a straddle is closer to zero, its gamma is higher. Gamma reflects how quickly the straddle’s sensitivity to changes in the underlying asset – i.e. the delta – will change. The trend strategy has no intra-month gamma, as once the position is set it remains static until the next rebalance.

As we generally expect the straddles struck longer ago to be deeper in the money than those struck more recently, we would also expect them to have lower gamma.

This also serves to nicely connect trend speed with the length of the lookback window. Shorter lookback windows are associated with trend models that change signals more rapidly while longer lookback windows are slower. Given that a total return trend signal can be thought of as the average of daily log returns, we would expect a longer lookback to react more slowly to recent changes than a shorter lookback because the longer lookback is averaging over more data.

But if we think of it through the lens of options – that the shorter lookback is coarsely replicating the delta of a straddle struck more recently – then the ideas of speed and gamma become linked.

Source: Tiingo and DiscountOptionData.com. Calculations by Newfound Research.

The Straddle Strategy as an Insurance Policy

One of the key differences between the trend strategy and the straddle is that the straddle has features that act as insurance against price reversals. As an example, consider a case where the trend strategy has a positive signal. To first replicate the payoff, the straddle strategy buys an in-the-money call option. This is the first form of insurance, as the total amount this position can lose is the premium paid for the option, while the trend strategy can lose significantly more.

The straddle strategy goes one step further, though, and would also buy a put option. So not only does it have a fixed loss on the call if price reverses course, but it can also profit if it reverses sufficiently.

One way to model the straddle strategies, then, is as insurance policies with varying deductibles. There is an up-front premium that is paid, and the strategy does not pay out until the deductible – the distance that the option is struck in the money – is met.

When the deductible is high – that is, when the trend is very strong in either direction – the premium for the insurance policy tends to be low. On the other hand, a strategy that purchases at the money straddles would be equivalent to buying insurance with no deductible.

Source: Tiingo and DiscountOptionData.com. Calculations by Newfound Research.

On average, the 3-month straddle strategy pays annual premiums of about 14% for the benefit of only having to wait for a price reversal of 6% before protection kicks in. Toward the other end of the spectrum, the 12-month strategy has an annual average premium of under 6% with a 16% deductible.

We can also visualize how often each straddle strategy pays higher premiums by looking at the deltas of the straddles over time. When these values deviate significantly from +1 or -1, then the straddle is lowering its insurance deductible in favor of paying more in premium. When the delta is nearly +1 or -1, then the straddle is buying higher deductible insurance that will take a larger whipsaw to payout.

The charts below show the delta over time in the straddle strategies vs. the trend allocation for 3-, 6-, and 12-month lookback windows.

There is significant overlap, especially as trends get longer. The differences in the deltas in the 3-month straddle model highlight its tradeoff between lower deductibles and higher insurance premiums. However, this leads it to be more adaptive at capitalizing on equity moves in the opposite direction that lead to losses in the binary trend-following model.

Source: Tiingo and DiscountOptionData.com. Calculations by Newfound Research.

The chart below shows the annualized performance of the straddle strategies when they underperform trend following (premium) and the annualized performance of the straddle strategies when they outperform trend following (payout). As the lookback window increases, both of these figures generally decline in absolute value.

Source: Tiingo and DiscountOptionData.com. Calculations by Newfound Research.

Even though we saw previously that the 3-month straddle strategy had the highest annual premium, its overall payout when it outperforms trend following is substantial. The longer lookbacks do not provide as much of a buffer due to their higher deductible levels, despite their lower premiums.

When the naïve trend strategy is right, it captures the full price change with no up-front premium. When it is wrong, however, it bears the full brunt of losses.

With the straddle strategy, the cost is paid up front for the benefit to not only protect against price reversals, but even potentially profit from them.

As a brief aside, a simpler options strategy with similar characteristics would be to buy only either a call option or put option depending on the trend signal. This strategy would not profit from a reversion of the trend, but it would cap losses. Comparing it to the straddle strategies highlights the cost and benefit of the added protection.

Source: Tiingo and DiscountOptionData.com. Calculations by Newfound Research.

Buying only puts or calls generally helped both of the strategies shown in the chart. This came in reduced premiums over a time period when trimming premiums whenever possible paid off, especially for the 12-month lookback strategy. However, there are some notable instances where the extra protection of the straddle was very helpful, e.g. August 2011 and late 2014 for the 3-month lookback strategy and March 2020 for both.

Despite the similarities between the options and trend strategies, this difference in when the payment is made – either up-front in the straddle strategy or after-the fact in whipsaw in the trend following strategy – ends up being the key differentiator.

The relative performance of the strategies shows that investors mostly benefitted over the past 15 years by bearing this risk of whipsaw and large, sudden price-reversals. However, as the final moths of data indicates, option strategies can provide benefits that option-like strategies cannot.

Ultimately, the choice between risks is up to investor preferences, and a diversified approach that pairs strategies different convex strategies such as trend following and options is likely most appropriate.

Conclusion

The convex payoff profile of trend following strategies naturally lends itself to comparative analysis with option strategies, which also have a convex payoff profile. In fact, we would argue – as we have many times in the past – that trend following strategies coarsely replicate the delta profile of option straddles.

In this commentary, we sought to make that connection more explicit by building option straddle strategies that correspond to a naïve trend following strategies of varying lookback lengths.

While the trend following approach has no explicit up-front cost, it risks bearing the full brunt of sudden and large price reversals. With the straddle-based approach, an investor explicitly pays an up-front premium to insure against these risks.

When evaluated through the lens of an insurance policy, the straddle strategy dynamically adjusts its associated premium and deductible over time. When trends are strong, for example, premiums paid tend to be lower, but the cost is a higher deductible. Conversely, when trends are flat, the premium is much higher, but the deductible is much lower.

We found that over the 2005-2020 test period, the cost of the option premiums exceeded the cost of whipsaw in the trend strategies in almost all cases. That is, until March 2020, when a significant and sudden market reversal allowed the straddle strategies to make up for 15 years of relative losses in a single month.

As we like to say: risk cannot be destroyed, only transformed. In this case, the trend strategy was willing to bear the risk of large intra-month price reversals to avoid paying any up-front premium. This was a benefit to the trend investor for 15 years. And then it wasn’t.

By constructing straddle strategies, we believe that we can better measure the trade-offs of trend following versus the explicit cost of insurance. While trend following may approximate the profile of a straddle, it sacrifices some of the intra-month insurance qualities to avoid an up-front premium. Whether this risk trade-off is ultimately worth it depends upon the risks an investor is willing to bear.

Why Trend Models Diverge

By Corey Hoffstein

On March 9, 2020

In Risk & Style Premia, Trend, Weekly Commentary

This post is available as a PDF download here.

Summary

During the week of February 23^rd, the S&P 500 fell more than 10%.
After a prolonged bullish period in equities, this tumultuous decline caused many trend-following signals to turn negative.
As we would expect, short-term signals across a variety of models turned negative. However, we also saw that price-minus-moving-average models turned negative across a broad horizon of lookbacks where the same was not true for other models.
In this brief research note, we aim to explain why common trend-following models are actually mathematically linked to one another and differ mainly in how they place weight on recent versus prior price changes.
We find that price-minus-moving-average models place the greatest weight on the most recent price changes, whereas models like time-series momentum place equal-weight across their lookback horizon.

In a market note we sent out last weekend, the following graphic was embedded:

What this table intends to capture is the percentage of trend signals that are on for a given model and lookback horizon (i.e. speed) on U.S. equities. The point we were trying to establish was that despite a very bearish week, trend models remained largely mixed. For frequent readers of our commentaries, it should come as no surprise that we were attempting to highlight the potential specification risks of selecting just one trend model to implement with (especially when coupled with timing luck!).

But there is a potentially interesting second lesson to learn here which is a bit more academic. Why does it look like the price-minus (i.e. price-minus-moving-average) models turned off, the time series momentum models partially turned off, and the cross-over (i.e. dual-moving-average-cross) signals largely remained positive?

While this note will be short, it will be somewhat technical. Therefore, we’ll spoil the ending: these signals are all mathematically linked.

They can all be decomposed into a weighted average of prior log-returns and the primary difference between the signals is the weighting concentration. The price-minus model front-weights, the time-series model equal weights, and the cross-over model tends to back-weight (largely dependent upon the length of the two moving averages). Thus, we would expect a price-minus model to react more quickly to large, recent changes.

If you want the gist of the results, just jump to the section The Weight of Prior Evidence, which provides graphical evidence of these weighting schemes.

Before we begin, we want to acknowledge that absolutely nothing in this note is novel. We are, by in large, simply re-stating work pioneered by Bruder, Dao, Richard, and Roncalli (2011); Marshall, Nguyen and Visaltanachoti (2012); Levine and Pedersen (2015); Beekhuizen and Hallerbach (2015); and Zakamulin (2015).

Decomposing Time-Series Momentum

We will begin by decomposing a time-series momentum value, which we will define as:

We will begin with a simple substitution:

Which implies that:

Simply put, time-series momentum puts equal weight on all the past price changes¹ that occur.

Decomposing Dual-Moving-Average-Crossover

We define the dual-moving-average-crossover as:

We assume m is less than n (i.e. the first moving average is “faster” than the second). Then, re-writing:

Here, we can make a cheeky transformation where we add and subtract the current price, P_t:

What we find is that the double-moving-average-crossover value is the difference in two weighted averages of time-series momentum values.

Decomposing Price-Minus-Moving-Average

This decomposition is trivial given the dual-moving-average-crossover. Simply,

The Weight of Prior Evidence

We have now shown that these decompositions are all mathematically related. Just as importantly, we have shown that all three methods are simply re-weighting schemes of prior price changes. To gain a sense of how past returns are weighted to generate a current signal, we can plot normalized weightings for different hypothetical models.

For TSMOM, we can easily see that shorter lookback models apply more weight on less data and therefore are likely to react faster to recent price changes.
PMAC models apply weight in a linear, declining fashion, with the most weight applied to the most recent price changes. What is interesting is that PMAC(50) puts far more weight on recent prices changes than the TSMOM(50) model does. For equivalent lookback periods, then, we would expect PMAC to react much more quickly. This is precisely why we saw PMAC models turn off in the most recent sell-off when other models did not: they are much more front-weighted.
DMAC models create a hump-shaped weighting profile, with increasing weight applied up until the length of the shorter lookback period, and then descending weight thereafter. If we wanted to, we could even create a back-weighted model, as we have with the DMAC(150, 200) example. In practice, it is common to see that m is approximately equal to n/4 (e.g. DMAC(50, 200)). Such a model underweights the most recent information relative to slightly less recent information.

Conclusion

In this brief research note, we demonstrated that common trend-following signals – namely time-series momentum, price-minus-moving-average, and dual-moving-average-crossover – are mathematically linked to one another. We find that prior price changes are the building blocks of each signal, with the primary differences being how those prior price changes are weighted.

Time-series momentum signals equally-weight prior price changes; price-minus-moving-average models tend to forward-weight prior price changes; and dual-moving-average-crossovers create a hump-like weighting function. The choice of which model to employ, then, expresses a view as to the relative importance we want to place on recent versus past price changes.

These results align with the trend signal changes seen over the past week during the rapid sell-off in the S&P 500. Price-minus-moving-average models appeared to turn negative much faster than time-series momentum or dual-moving-average-crossover signals.

By decomposing these models into their most basic and shared form, we again highlight the potential specification risks that can arise from electing to employ just one model. This is particularly true if an investor selects just one of these models without realizing the implicit choice they have made about the relative importance they would like to place on recent versus past returns.