by Daniel Leveau, VP Investor Solutions & Peter Huber, Investment Writer
Discover how to build three systematic investment strategies, covering the asset classes equities, fixed income and commodities.
We hosted our Quant Insights panel discussion in early 2022, where SigTech’s Stuart Broadfoot remarked on the essential volatility of markets, noting that ‘history never repeats itself perfectly and unforeseen events occur.’ Whether seeking to manage short term uncertainty or to gain a long term advantage with a buy and hold approach, successfully navigating the financial markets requires a well-structured investment strategy. Systematic, data-driven approaches to investment offer such coherence; able to generate more consistent market returns via their rational, objective, rules-based approach to investment. Systematic investment strategies help an investment manager determine the ideal asset allocation for a portfolio with the objective of optimizing future results. Furthermore, systematic investment strategies enable investors to mitigate the distortionary effects of emotion, which turbulent market environments may otherwise potentiate.
In this article, we present three investment strategies – covering the asset classes equities, fixed income and commodities – that are drawn from our extensive library of over 300 fully customizable quantitative investment strategies and functionalities available on SigTech’s quant technology platform. You can also watch a demo video for each strategy.
Multi-Factor Equity Strategy
To create our multi-factor equity strategy we begin by defining the stock market index S&P 500 as the investment universe and a monthly rebalancing frequency (as per end of month).
The next step is – using one of 300+ prebuilt investment strategies and functionalities available on our quant technology platform – to construct a reinvestment strategy for each stock in the investment universe, taking historical additions and deletions to the S&P 500 index into consideration. We also customize how we treat historical corporate actions such as dividend payments and rights issues.
We then select the various risk factors used in our analysis to assess the attractiveness of each stock. The risk factors we use are quality, value, and momentum.
Quality factors
- Net debt-to-EBIT (evaluate financial leverage)
- Cash ratio (evaluate liquidity)
- Return on capital (evaluate profitability)
- EBIT margin (evaluate profitability)
Value factors
- Price-to-book
Price-to-free-cash-flow
- Price-to-earnings
- Dividend yield
Momentum factor
- Six month price momentum
We calculate each individual factor at a stock level and then combine them – using their Z-scores – to get an aggregated value, quality and momentum score, respectively. Factor weights across the three risk factors are applied to these aggregate values to calculate a final composite score for each stock. The top 20% of stocks are selected for inclusion in our portfolio and each of the selected stocks are given an equal-weight on each rebalancing date.
To assess the robustness of our strategy, we define a set of factor weights and perform a series of backtests across a specified date range for each iteration. The results are tabulated in our customizable analytics report.
The best performing strategy is the one applying a 50% weight to value and 50% to quality. It handily outperforms the market capitalization-weighted S&P 500 index and exhibits a Sharpe ratio > 2 and an annualized return of approx. 30%.
Our interactive performance plot allows for further analysis of the strategy.
We continue by analyzing the portfolio’s risk profile through its exposure to the Fama-French factors. We begin the analysis by loading the index returns of the factors; market (MKT), size (SML), and value (HML). A multivariate regression of each stock’s returns against the three Fama indices is performed, and the portfolio’s rolling exposure to the three factors over time is calculated.
Generating a factor exposure report provides greater detail of the various strategies’ exposures, decompositions and distributions relative to an equally weighted universe over time.
Brent Crude Oil Strategy
Our Brent crude oil strategy uses a trading signal based on the PMI manufacturing index and is hedged with an options collar to reduce volatility.
As an initial step, we use the rolling futures strategy building block to create a customized historical time series that rolls crude oil futures over time according to our specifications. We specify a roll strategy that rolls half of the contracts six days before expiry, and the remaining five days before expiry.
To avoid re-investing PnL we set target weights from initial cash and configure the initial cash to USD 100k for our strategy. Matching notional exposures allows for the hedging of future positions. By not re-investing PnL we ensure that USD 100k worth of futures are hedged with USD 100k worth of options.
To create the trading signal, we compute a momentum signal using the monthly US manufacturing PMI. The direction of the strategy’s exposure to the Brent crude oil futures is a function of the signal’s value crossing the two month exponentially weighted moving average; when the actual PMI value rises above the average, the strategy goes long and when it falls below the average, it goes short.
Whilst the strategy is profitable, returns are volatile with the strategy suffering large drawdowns. To improve the strategy’s return profile, we apply a protective options collar strategy as an overlay.
To create the overlay strategy, we again make use of the reinvestment strategy building block and define a basket of tradable options. Baskets of options are created on each rebalance date, each specifying the number of units to trade.
To construct the protective collar, options are struck 10% out-of-the money with three month maturities. A fixed spot notional target of USD 100K is set to match the portfolio exposure taken in the futures strategy. The overlay strategy will trade the collar in the same direction as the Brent futures; when the strategy shorts oil futures, the overlay strategy shorts options, and vice versa.
SigTech’s performance report compares our two backtests – i.e. the strategy with and without the options overlay strategy – across key metrics (e.g. annualized excess return, max drawdown, volatility). Various graphical visualizations such as rolling plots can be generated to present the results. As shown by the statistics below, applying the overlay strategy to the Brent crude oil strategy markedly improves the risk-return profile.
OTR Bond Strategy
The rolling bond strategy building block allows you to define a strategy’s country, currency, and start date. Bonds are selected by specifying tenor and run type. Using this template you can, for instance, systematically roll US treasuries, maintaining exposure at a particular tenor.
With this building block we build two preliminary strategies.
- A rolling on-the-run 10Y treasuries strategy
- A rolling first-off-the-run 5Y treasuries strategy
The SigTech backtesting engine facilitates the analysis at any historical point in time, both with respect to the orders placed and to positions held. Backtesting these strategies and comparing their performance reveals that our rolling on-the-run 10Y treasuries strategy would have produced higher market returns than our first-off-the-run 5Y strategy if traded over the specified window.
The basket strategy building block allows the bundling of financial instruments and tradable strategies. Using this building block we can define start date, currency, weights, rebalance dates, and rebalance frequency.
Our basket strategy contains the following three constituents:
Our existing rolling on-the-run 10Y treasuries strategy
- A Japanese Bond rolling future strategy
- A US Treasury rolling future strategy
We are long on the first instrument, short on the second instrument and long on the final instrument. Rebalancing occurs at the end of the month. The two rolling future strategies are created with SigTech’s rolling future strategy building block, following a similar process to the one used to create our rolling bond strategies.
We then backtest this strategy and compare its market returns with those of our rolling on-the-run 10Y treasuries strategy. The performance report provides a table of key metrics (e.g. annualized excess returns, volatility, max drawdown, etc.). The basket strategy outperforms our bond strategy, with an annualized excess return of 1.22% compared to 0.84%.
Conclusion
Hedge funds, pension funds, mutual funds and exchange traded fund (ETF) providers increasingly recognize the need to systematize their analytics and investment processes.The presence of these processes and the technological infrastructure on which they depend can speed up their investment decision process.
SigTech offers a future-proof quant technology platform for fund managers seeking to apply a quantitative trading infrastructure. We eliminate the expensive upfront costs of infrastructure build-out and provide everything you need to construct, backtest, optimize, and deploy your trading strategies. Our research environment combines a wealth of clean, curated, and operationally-ready datasets with a comprehensive library of over 300 pre-built and fully customizable investment strategies.
Please contact us to discuss how we can help you accelerate your investment process.
Disclaimer
This content is not, and should not be construed as financial advice or an invitation to purchase financial products. It is provided for information purposes only and is subject to the terms and conditions of our disclaimer which can be accessed here.
