Systematic trading is often synonymous with short-term trading frequencies in the retail quant trading space. Daily and intraday strategies tend to receive the bulk of the community's attention. The popularity of systematic cryptocurrency trading has put a further emphasis on short term trading methods.
Higher frequency strategies may be appropriate for active traders who wish to enhance returns over a short-term horizon. They may be less appropriate for those considering much longer term goals.
One such long-term investment goal is retirement planning. There are many quantitatively-minded retail traders who wish to control some or all of the decision making with regards their pension plan. However the constant daily monitoring of higher frequency systematic trading strategies may not be their primary focus. Instead they may desire to adapt their portfolios at a much lower frequency.
This is where the concept of systematic Tactical Asset Allocation comes in.
In this article we are going to introduce tactical asset allocation (TAA). We will define its investment approach—particularly as it relates to systematic methodology. We will discuss how TAA differs from both buy & hold and shorter-term strategies. Then we will present some advantages and disadvantages of the approach. To firm up our understanding of TAA we will consider a real-world example strategy.
What is Tactical Asset Allocation?
Tactical asset allocation is an investment approach that optimises for a long-term investment horizon with relatively infrequent portfolio rebalances.
Most TAA strategies are rebalanced once per month, but quarterly or even semi-annually rebalances are not uncommon.
TAA strategies invest in broad asset classes—typically equities, fixed income and cash-like securities. However the rise of ETFs has meant that more exotic asset classes are now included in TAA strategies, including real-estate, commodities and credit.
TAA is 'tactical' because at each rebalance period it is possible to increase allocations to asset classes that are expected to produce relative outperformance, while reducing allocations to those that may lead to relative underperformance. This makes TAA a dynamic asset allocation strategy.
TAA is attractive because it mitigates the effect of daily variance and instead exploits well-known, academically robust market factors to produce outsized returns or reduced risk. Such factors include anomalies such as 'momentum', 'value' and 'quality'.
While TAA has been an extremely prevalent active discretionary strategy it is gaining in popularity as a systematic, quantitative-focused strategy as well. This removes all decision making from the execution allowing a straightforward implementation.
TAA is a great place to start for those who are new to systematic trading. Certain strategies require little experience of coding since they can be easily backtested in spreadsheet software. The investment rules are often easy to understand and implement.
Despite the initial ease of TAA it is possible to create complex strategies. Since most TAA strategies are implemented via long-only investment of ETFs, a vast array of complex portfolios can be generated. Adding a quantitative portfolio construction and risk management methodology to TAA signals can lead to a sophisticated, robust long-term investment strategy.
Advantages of Systematic Tactical Asset Allocation
There are various attributes that make tactical asset allocation attractive for the long-term quantitative retail trader:
- Codification - Signal rules are straightforward to codify, often requiring a simple 'flowchart' approach. These rules are unambiguous and can be implemented straightforwardly in nearly any backtesting framework.
- Transaction Costs - TAA strategies often rebalance at most once per month. If the dynamic weights calculated for each asset class are slow-moving this can vastly reduce transaction costs when compared to a more frequently traded strategy. This can produce a staggering difference in CAGR—and thus effectiveness of the retirement plan—over a typical long-term investment horizon.
- ETF Instruments - Most TAA strategies rely on investing in highly liquid Exchange Traded Fund (ETF) instruments as a proxy for the asset classes being invested in. ETFs exist for nearly any asset class imaginable and as such it can be very straightforward for the retail quant to begin implementing a portfolio.
- Long Only - Most TAA strategies prescribe an unlevered long-only approach. This makes such strategies accessible to many individuals that would otherwise be unable to trade utilising margin.
- Meta Strategies - Tactical asset allocation strategies can be easily combined into a 'meta portfolio' by implementing an additional layer of 'meta weights' to a collection of individual TAA strategies. Such meta strategies allow almost infinite customisation for the individual investor's personal risk/reward preferences.
Disadvantages of Systematic Tactical Asset Allocation
Tactical Asset Allocation is not without its disadvantages:
- Forecasting - TAA approaches implicitly assume the ability to forecast movements of broad asset classes over the short to medium term. This means exploiting factors such as momentum, value and quality. Hence there is a need to research approaches to forecasting such factors or outsourcing the forecasting to others who have developed strategy rules.
- Timing Luck - TAA strategies are particularly sensitive to timing luck, which is the empirically-observered dispersion of returns that can occur due to the chosen rebalance date of the strategy. For instance there can be substantial differences in long-term performance if the rebalance date is chosen mid-month as opposed to end-of-month.
- High Beta - Since TAA strategies are long-only and often have a substantial equities component, without additional portfolio construction overlays they often have a high beta component and thus are sensitive to overall equities market moves.
- Transaction Costs - Certain TAA strategies make use of 100% allocations to certain asset classes in some months. This can mean a complete liquidation of a portfolio on a monthly basis for highly dynamic strategies. There are substantial costs associated with such turnover. However these costs can be minimised by risk overlays that enforce minimum allocations to certain classes in an attempt to reduce the turnover.
Systematic Tactical Asset Allocation Example - Dual Momentum
Perhaps the most famous tactical asset allocation strategy within the retail quant trading community is Gary Antonacci's Dual Momentum GEM model.
The strategy combines the concepts of relative strength and absolute momentum (aka cross-sectional momentum and time-series momentum) in an attempt to produce higher returns with a lower overall volatility and reduced drawdowns.
The standard rules of the strategy are straightforward. One approach utilises the SPY (S&P500 Equities), VEU (FTSE All-World ex-US Equities), BIL (SPDR/Bloomberg/Barclays 1-3 Month T-Bills) and BND (Vanguard Total Bond Market) ETFs as proxies for each asset class.
At the first business day of each month:
1) Determine whether the 12-month returns of SPY exceed those of VEU.
2) If so, then determine whether the 12-month returns of SPY exceed those of BIL. If so go long SPY, else go long BIL.
3) Conversely if VEU 12-month returns exceed SPY returns, determine whether VEU returns exceed those of BIL. If so go long VEU, else go long BIL.
It is clear that such rules can be very straightforwardly implemented in spreadsheet software and can be easily updated once per month. Its rules are completely codified and require no discretion on the part of the investor. Each of the ETFs are highly liquid and thus will have low transaction costs, which combined with the strategy's historically low turnover make it a relatively cheap strategy to invest in. Lastly it exploits an extremely well-known and robust market anomaly in the momentum factor.
Despite these advantages is does possess an intrinsic drawback. The rules above ensure that the strategy is 100% allocated to a particular asset class at any one time and, depending upon the chosen initiation date, has historically spent approximately 70% of its time in stocks. This high concentration in specific asset classes may be difficult for the traditional quant to cope with. The traditional quant is likely to be used to significant diversification through more quantitative portfolio construction methodologies.
Nevertheless it is clear that TAA strategies can be simple, easy to implement and can exploit well known market anomalies to produce excess returns above a more traditional buy & hold approach.
Implementation in QSTrader
For signal generation it includes fixed-weight (strategic asset allocation) and dynamic-weight (tactical asset allocation) capabilities. These weights can then be fed into a portfolio construction framework to produce a list of rebalance trades. Currently our development version supports weekly and end-of-month rebalances.
We will be providing more examples of TAA strategies as the development of QSTrader matures in the next few months. If you would like to be involved in the development of QSTrader please contact us at firstname.lastname@example.org and we will send out an invite to our QSTrader Slack development channel.
In subsequent articles we will go into more depth about implementation of such strategies.
In order to investigate the concentration risk and potentially large turnover inherent to the TAA approach we will be producing our own research, which will include investigation into strategy performance enhancement via the addition of robust portfolio construction and risk management techniques.
If you have any questions about systematic tactical asset allocation and how to get started please feel free to email us at email@example.com and we will do our best to provide you with some guidance.