Adaptive Risk Parity: How AI Transforms the All-Weather Portfolio

4 Minute Read

The first half of 2023 saw financial markets grappling with the consequences of monetary policy normalization and geopolitical challenges. Investors entered 2023 on the back of the worst year for the S&P 500 since the 2008 financial crisis, and have had to contend with a banking crisis, sticky inflation, and interest rate hikes (and pauses) from the Federal Reserve. The S&P 500 Index is up 16.89% for the year as of the end of June, thanks in no small part to the AI Boom.[1] However, investor sentiment remains cautious as lingering recession fears and questions raised over US national debt signal an uncertain market environment ahead.

In response to increasingly volatile market conditions, risk parity strategies are becoming a popular approach to managing investments and seeking more stable risk-adjusted returns. Initially coined by Edward Qian in 2005[2], the term risk parity refers to portfolios that attempt to equalize risk by diversifying across different types of assets such as stocks, bonds, commodities, and other alternatives. These diversified portfolios aim to spread investment risk across multiple asset classes and take advantage of various market opportunities through different business cycles and market environments. One limitation of traditional risk parity strategies is that they rely on a static rule-based methodology when determining risk allocation among different asset classes, thus rendering them liable to underperformance when the diversification effect fails.

2022 was a particularly difficult year for risk parity strategies because fixed income did not function as a defensive asset class – when the assumption of a negative stock-bond correlation fails and the risk diversification effect of bonds no longer apply, static risk parity strategies are too inflexible to react to changing market regimes. The Qraft AI-Powered All-Weather Portfolio aims to address these limitations by applying artificial intelligence (AI) to create a more dynamic and adaptive risk parity strategy.

Qraft AI-Powered All-Weather Tilt: Methodology

In order to strike a balance between risk mitigation and upside capture, Qraft employs a tactical risk budgeting strategy that adjusts the level of risk in a portfolio in accordance with market conditions or specific risk goals instead of remaining static. Qraft’s proprietary AI engine considers the risk profile of each investment and determines their respective portfolio allocations based on predefined risk limits as well as the nowcasted market regime. For example, the All-Weather Tilt portfolio may opt to overweight riskier asset classes such as equities when the market is stable, and vice versa. This method incorporates market dynamics into the balance between risk and potential returns when deciding how much to invest in each asset.

Qraft’s proprietary AI engine utilizes a deep neural network to actively generate signals by analyzing macro variables such as interest rates and economic growth, among others, and their impact on each asset class. Qraft’s AI engine is constantly evolving based on new market trends and optimization of risk-adjusted returns and adjusts its allocation monthly.

Qraft AI-Powered All-Weather Tilt: Performance Review

2023 has been fruitful so far for the Qraft AI-Powered All-Weather Tilt Portfolio. In comparison to a comparable ETF strategy, RPAR Risk Parity ETF (Ticker: RPAR), the All-Weather Tilt recorded a YTD return of 9.47% (vs 4.49%) as of 30 Jun 2023. Over the same period, the All-Weather Tilt recorded a 109.83 Up Capture Ratio and 57.13 Down Capture Ratio, highlighting the strategy’s ability to navigate changing market regimes far more dynamically than static portfolios. Additionally, the Qraft AI-Powered All-Weather Tilt Portfolio has maintained a lower volatility of 12.27 (vs 14.95%) against RPAR in H1 2023 earning a Sharpe ratio of 1.12 (vs 0.31).[3]

[Figure 1] YTD Investment Growth (1 Jan 2023 - 30 Jun 2023)

[Figure 2] YTD Investment Drawdown (1 Jan 2023 - 30 Jun 2023)

[Figure 3] Investment Growth Since 2020 (1 Jan 2020 - 30 Jun 2023)

[Figure 4] Investment Drawdown Since 2020 (1 Jan 2020 - 30 Jun 2023)

In 2022, a year where risk parity strategies, in general, struggled due to the positive correlation between equity and bond returns, Qraft’s AI-Powered All-Weather Tilt managed to reduce drawdowns better than the RPAR strategy (Maximum drawdown: All Weather Tilt -24.86%, RPAR -30.07%).[4]

Catching the June Stock Rally

Qraft’s AI-Powered All-Weather Tilt Portfolio has proven to be successful in managing the S&P 500's highs and lows over the past few months. This strategic approach towards risk allocation allowed the portfolio to adopt a risk-on approach when the market conviction was high, while still providing protection against market downturns and volatility through diversifying across asset classes. [Figure 5] shows the All-Weather Tilt’s equity exposure in relation to the S&P 500 Index[5] and displays the portfolio’s ability to dynamically tilt asset class allocation in response to changing market environments.

[Figure 5] Portfolio Asset Allocation and S&P 500 Total Return (1 Jan 2023 - 30 June 2023)[6]

As shown, the All-Weather Tilt Portfolio demonstrated the ability to tactically adjust its equity exposure, first capitalizing on rising markets in January before switching to other asset classes after the S&P 500 suffered lagging returns in February. Over the past few months, the All-Weather portfolio gradually increased its allocation to equities, while underweighting on fixed incomes and commodities.

Conclusion 

In summary, our AI-Powered All-Weather Portfolio has demonstrated its effectiveness in navigating the turbulent global financial landscape over the past few years. By leveraging artificial intelligence, this portfolio has enhanced the traditional risk parity approach, offering a more dynamic and adaptive investment strategy. The AI’s ability to adapt to changing market conditions has allowed it to capture opportunities and mitigate risks, delivering superior returns while maintaining portfolio stability. Over the long run, it has proven to be a successful example of how AI can optimize diversification to maximize returns, positioning investors for long-term success in an ever-evolving economic environment.

[1] Morningstar Direct, 30 June 2023.

[2] Qian, E (September 2005). “Risk parity portfolio: Efficient portfolios through true diversification”.

[3] Morningstar Direct, 0 bps transaction cost.

[4] Morningstar Direct, 0 bps transaction cost.

[5] Morningstar Direct, 30 June 2023.

[6] Portfolio is rebalanced on the first business day of each month.

Featured