Amit Das
Over the past few decades, the US has exhibited extremely resilient growth with a handful of sharp bear markets along the way. However, the positive trend has been consistent and recoveries after bear markets have been typically quick. We can use the S&P 500 to illustrate this. Following the initial shock of the COVID-19 pandemic, the S&P 500 lost about 30% of its market value. However, it began to steadily increase again within a month. Within six months it had made up the lost ground entirely. By year-end, it was up over 10% from the start of the pandemic.
In this insight, we examine whether news has the potential to serve as an early indicator for bear markets in the US.
How cyclical is news?
Prior LSEG research has examined how LSEG’s Machine Readable News and News Analytics feeds can be successfully used to predict stock-specific actions. In the case of stock-specific news, we tend to observe high sparsity. The result of this is that some of the strongest news signals, on a day-to-day basis, come “out of the blue”. They are on stocks which either did not have much news about them in the prior weeks (or even months) or on companies that maintain a steady stream of low polarity news. The latter is typically true for some of the largest companies that dominate public attention, such as Tesla (TSLA) and Amazon (AMZN).
The goal here is to instead leverage news to predict broader market trends. Looking from this perspective, news is not at all sparse. Tens of thousands of articles are produced and recorded every trading day. However, a new issue emerges as we move from looking at news for specific companies to aggregates across the US market. We find that we replace the challenge of sparsity with that of serial correlation. As can be seen in figure 1, daily aggregate news sentiment within the US exhibits high levels of autocorrelation, across even relatively long lags.
Figure 1: Lagged autocorrelation of daily aggregate news sentiment in the US. Daily aggregates are calculated by summing up the article-specific sentiment scores for all articles relevant to US listed stocks.
This seems to present a real challenge to our stated aim – to use aggregated news signals to predict sharp changes in the US market. To achieve this, we require an indicator that is as close to memory-less as possible. Such an indicator would have a better chance of sensing market shifts in time.
Fortunately, a simple adjustment to our daily aggregate score can be made to address this challenge. To generate a more dynamic signal, let us instead look at the daily change in aggregated news sentiment. Figure 2 shows us that this indicator does not exhibit any significant autocorrelation after the first day. The negative first day autocorrelation is indicative of some amount of mean-reverting behaviour. Assuming the financial and news market conditions are broadly stable, we expect the daily aggregated news sentiment to hover around some average value. As such, if it veers too far from average the next day, we expect to move back toward average the following day.
Figure 2: Lagged autocorrelations of the change in daily aggregate US news sentiment.
News as an indicator
To leverage our change in news sentiment as an indicator, it is useful to make one adjustment based on the 1-day autocorrelation observed. Specifically, it is good to average the indicator across a handful of look-back days. This provides some robustness to the back-and-forth behaviour outlined above. For this insight, we use a look-back of five days. An alternative approach could be to only consider high values of change in daily sentiment, e.g., two standard deviations below the average daily change.
Figure 3 shows the daily closing price of SPY across two of the most recent bear markets in the US. The first is during the start of the pandemic. The second is the beginning of the more protracted bear market that started in 2021 and extended through to 2022. The shading in the background shows our sentiment change detector. As above, this relies on the change in daily aggregate sentiment with five-day simple smoothing.
In both instances we see a multi-day negative signal which is unprompted by any movement in the markets. In both cases, this prolonged negative signal precedes a significant market downturn.
Source information
Figure 3: SPY price layered against our indicator. The indicator is constructed by taking the five-day moving average of the change in daily US news sentiment. Red indicates a negative value, green indicates positive.
Investors in the US market tend to operate in a bullish environment. As such, both systematic and discretionary traders can end up operating in a manner that is most effective during bull runs. Systematic traders will tend to train learning models and tune hyper-parameters on data primarily taken from bull markets. Discretionary traders will have learned habits in a similar way. The use of news as an early warning system for impending breaks to the status quo could aid all investors to be as nimble as required.
Legal Disclaimer
Republication or redistribution of LSE Group content is prohibited without our prior written consent.
The content of this publication is for informational purposes only and has no legal effect, does not form part of any contract, does not, and does not seek to constitute advice of any nature and no reliance should be placed upon statements contained herein. Whilst reasonable efforts have been taken to ensure that the contents of this publication are accurate and reliable, LSE Group does not guarantee that this document is free from errors or omissions; therefore, you may not rely upon the content of this document under any circumstances and you should seek your own independent legal, investment, tax and other advice. Neither We nor our affiliates shall be liable for any errors, inaccuracies or delays in the publication or any other content, or for any actions taken by you in reliance thereon.
Copyright © 2025 London Stock Exchange Group. All rights reserved.
The content of this publication is provided by London Stock Exchange Group plc, its applicable group undertakings and/or its affiliates or licensors (the “LSE Group” or “We”) exclusively.
Neither We nor our affiliates guarantee the accuracy of or endorse the views or opinions given by any third party content provider, advertiser, sponsor or other user. We may link to, reference, or promote websites, applications and/or services from third parties. You agree that We are not responsible for, and do not control such non-LSE Group websites, applications or services.
The content of this publication is for informational purposes only. All information and data contained in this publication is obtained by LSE Group from sources believed by it to be accurate and reliable. Because of the possibility of human and mechanical error as well as other factors, however, such information and data are provided "as is" without warranty of any kind. You understand and agree that this publication does not, and does not seek to, constitute advice of any nature. You may not rely upon the content of this document under any circumstances and should seek your own independent legal, tax or investment advice or opinion regarding the suitability, value or profitability of any particular security, portfolio or investment strategy. Neither We nor our affiliates shall be liable for any errors, inaccuracies or delays in the publication or any other content, or for any actions taken by you in reliance thereon. You expressly agree that your use of the publication and its content is at your sole risk.
To the fullest extent permitted by applicable law, LSE Group, expressly disclaims any representation or warranties, express or implied, including, without limitation, any representations or warranties of performance, merchantability, fitness for a particular purpose, accuracy, completeness, reliability and non-infringement. LSE Group, its subsidiaries, its affiliates and their respective shareholders, directors, officers employees, agents, advertisers, content providers and licensors (collectively referred to as the “LSE Group Parties”) disclaim all responsibility for any loss, liability or damage of any kind resulting from or related to access, use or the unavailability of the publication (or any part of it); and none of the LSE Group Parties will be liable (jointly or severally) to you for any direct, indirect, consequential, special, incidental, punitive or exemplary damages, howsoever arising, even if any member of the LSE Group Parties are advised in advance of the possibility of such damages or could have foreseen any such damages arising or resulting from the use of, or inability to use, the information contained in the publication. For the avoidance of doubt, the LSE Group Parties shall have no liability for any losses, claims, demands, actions, proceedings, damages, costs or expenses arising out of, or in any way connected with, the information contained in this document.
LSE Group is the owner of various intellectual property rights ("IPR”), including but not limited to, numerous trademarks that are used to identify, advertise, and promote LSE Group products, services and activities. Nothing contained herein should be construed as granting any licence or right to use any of the trademarks or any other LSE Group IPR for any purpose whatsoever without the written permission or applicable licence terms.
