10 new paradigms to boost returns in 2024

The markets may end up closing the year with substantial gains, but 2023 was not without dramatic turns. From the wars in Ukraine and Gaza to global inflation, observers saw a rise in uncertainty that stretched legacy approaches to risk management, and emphasized a pressing need for timely investment insights. 2024 may not be smooth sailing either: beside the rumblings of a possible recession, the world is bracing for a polarized US election cycle, and geopolitical tensions may further escalate in the new year.

Traditionally, risk was manageable, quantified by likelihood and impact, but the nature of risk has shifted. As the speed of information increased, so did the speed of risk.

Situations which would escalate progressively may now explode overnight, with far-reaching consequences on portfolio composition, hedging requirements, and even strategic outlook. For the valuation of many tech companies, there was a day before and a day after the release of chatGPT. October 7 shook the world, but also sent crude oil and gold prices surging. Data accelerates, and signals erode quickly. To remain in control, investors need reliable and timely data points they can easily consume, quickly onboard, and build upon in their decision making process.

To address these needs, data vendors have produced a new generation of datasets which emphasize timeliness in delivery and ease of use. Leveraging quantitative approaches, but presented as straightforward factors, these new products provide a mosaic of indicators that can be quickly leveraged to gauge the impact of external events in real time, proxy lagging macroeconomic values, and even capture the mindset of human analysis as it happens. In this article, we present 10 behavioral improvements, powered by newly-released factors and research, that investors can embrace to better prepare for the expected volatility of 2024.

Know if and when the recession hits today, not 3 months later

A timely assessment of the current state of the business cycle can empower investors to make informed decisions, strategically position their portfolios, and seize opportunities that align with the dynamic shifts in the economic landscape. Traditional macroeconomic indicators, which are often lagging, fail to capture the nuanced and rapidly evolving trends.

The Economic Sentiment Indicator aggregates the views of corporate executives about the current and future prospects of the US economy. By detecting inflection points as they occur, the indicator acts as an early warning of regime changes, and delivers a more accurate assessment of the future state of the US economy. Updated daily, the indicator takes into account every earnings release to update its view of the economy. With up to 6 weeks of earnings season every quarter, the indicator effectively acts as a nowcast view of the US business cycle.

Stop paying the inflation tax

A timely assessment of inflation is crucial, especially during uncertain times, for effectively hedging inflation risks, reallocating capital to commodities, or reducing exposure to interest rate fluctuations. RavenPack's Inflation Nowcasting indicators utilize machine learning models to predict both US Core CPI and Headline CPI on a daily basis. These indicators offer a Backcast, estimating the previous month's inflation until official data is released, and a Nowcast, projecting inflation figures expected by month-end. Continuously updated until official figures are available, the Backcast provides a two-week advanced insight into the prior month's inflation. This early information aids in preparing for potential inflation surprises, allowing for real-time adjustments to portfolio risk exposure.

Backed by research: we show how to build an inflation-based asset allocation strategy using sentiment data and we illustrate that sentiment-based strategies outperform models that depend merely on past observed inflation values.

Keep a finger on the pulse of the US economy

The US economy continues to drive global markets, and obtaining reliable and timely updates on its evolution gives investors a competitive edge to navigate uncertainties, and fine-tune their strategies, especially when paired with inflation and other nowcasts. Combining high-frequency macro factors with refined sentiment scores minimizes noise, and enhances forecast accuracy. RavenPack's US GDP Nowcasting model, powered by proprietary machine learning, predicts daily US GDP growth rates: Backcasting Q–1, Nowcasting the current Q, and Forecasting Q+1 to Q+4. Outperforming benchmarks, this model significantly enhances predictions weeks before official figures are released.

Backed by research: this innovative model uses RavenPack sentiment to improve GDP predictions by up to 70%.

Stop being in the dark on China’s economy for two months

Chinese macroeconomic data for January and February is published jointly at the beginning of March, in accordance with the Lunar calendar. This leaves investors without a clear picture of the economy's trajectory at the start of the year. To address this, we developed a model to predict China’s GDP growth by blending traditional macro indicators with alternative data like electricity use, freight volumes, and sentiment analysis. The RavenPack China GDP Nowcasting model leverages proprietary machine learning models to predict, on a daily basis, the China GDP growth rate for the previous quarter Q–1 (Backcast), the current quarter Q (Nowcast), and the next Q+1, Q+2, Q+3, and Q+4 quarters (Forecasts).

Backed by research: the addition of RavenPack Analytics to the model improves performance in terms of the out-of-sample forecasting error by up to 9%, compared to a state-of-the-art nowcasting model, which uses only macroeconomic variables.

Use effective ESG signals to enhance portfolio performance

Avoiding high-controversy stocks and bonds enhances returns, minimizes risk and fosters investments in companies with stronger environmental compliance and more robust ethical standards. However, detecting negative news on a large scale poses challenges. RavenPack Controversy Scores offer both discretionary and quantitative investors streamlined daily reports on controversial events, for more informed decision-making.

Backed by research: we demonstrate the predictive power of sentiment analysis in earnings news at the sector level. Excluding the top 50% of companies with the greatest corporate controversy exposure enhanced results when allocating to the Top-3 sectors, while slightly reducing volatility and downside risks.

Leverage LLMs trained with quality data to refine equity trading

The rise of large language models has reignited investor focus on sentiment-based trading. While open-source models perform decently, their effectiveness hinges on quality training data. RavenPack optimized FinBERT—an open-source model trained on financial data—by incorporating RavenPack Annotations - a data set which contains entity-specific sentiment data per sentence. This convergence between powerful language models and quality training data enhances stock market movement predictions, paving the way to a new era of precision in finance.

Backed by research: when applied to back-testing six investment strategies across global markets, the fine-tuned FinBERT model increases the annualized returns by 47%.

Catch more untapped opportunities during earnings seasons

With hundreds of companies holding their earnings conference calls on the same day, it’s hard to keep up. RavenPack's Earning Intelligence factors tap news, earnings call transcripts, and even insider transactions, to construct high-performing factors that capture more information during and after earnings season. Discretionary and quantitative investors benefit from these factors by mitigating portfolio risk, identifying surprises that affect volatility, refining stock selection through timely rebalances, and incorporating guidance and market consensus into strategies.

Backed by research: 5 Ways NLP Can Help Uncover More Alpha During Earnings Season, now also for APAC.

Embrace cutting-edge approaches to monitor and measure innovation

Traditional methods of tracking innovative companies are resource-intensive. Data-driven approaches, employing NLP and textual data sources, offer efficient alternatives. These techniques swiftly identify firms aligned with emerging themes, empowering timely, informed investment decisions.

For instance, by leveraging RavenPack Job Analytics, we can quantify tech adoption in hiring to assess investment potential. Additionally, a news-driven strategy constructs equity portfolios around innovation, with media attention proving to be a robust predictor of firm performance across regions for mid- to large-size companies from 2011 to 2023.

Improve the timing of sector rotations

As we navigate market uncertainty, leveraging real-time narrative shifts through sentiment analysis can prove useful for profitable sector rotation. In a volatile landscape, understanding which industries lead and which lag becomes crucial. By tapping into cutting-edge sentiment analysis, we can gauge sector momentum and rebalance portfolios for optimal gains.

RavenPack's research indicates that harnessing earnings sentiment informs smarter sector allocation and rebalancing decisions. This innovative approach merges earnings news, macro data, and bonds to supercharge equity sector rotation.

Use News Sentiment to maximize Forex Futures returns

As we navigate market uncertainty, leveraging real-time narrative shifts through sentiment analysis can prove useful for profitable sector rotation. In a volatile landscape, understanding which industries lead and which lag becomes crucial. By tapping into cutting-edge sentiment analysis, we can gauge sector momentum and rebalance portfolios for optimal gains.

As foreign exchange markets grow more volatile due to macroeconomic news and geopolitics, sentiment becomes a key component in boosting signals for FX futures strategies. Notably, events like the Russian invasion of Ukraine and the recent inflation surge, which led to rising interest rates, have reshaped trade dynamics. New research demonstrates the power of news sentiment data in predicting currency performance and long-short strategies for G7 currency pairs.

Adaptability and data-driven strategies hold the key to success in a volatile market. Explore the RavenPack Factors Library, along with an extensive collection of research-backed insights. This comprehensive toolkit equips you to navigate market volatility with confidence, ensuring your investment decisions are grounded in real-time data and strategic foresight.