Products
Media Analysis
LSEG MarketPsych Analytics
Rich sentiment data tracking news and social media in real time
LSEG MarketPsych ESG Analytics
ESG data designed to drive better investment decisions
LSEG MarketPsych Transcript Analytics
Earnings calls analytics for investing, research and risk management
Prediction and Monitoring
StarMine Media Sentiment Model
30-day prediction model for global equities using sentiment
MarketPsych Event Pulse
An AI-driven investment news feed
Advanced Solutions
Scoring Engine for Natural Text
An NLP-as-a-service platform for financial text analytics
October 01, 2024

Is the Trend Still Your Friend?

Dear Readers,

Today we open with one of the earliest observations of human behavior in financial markets. More than 400 years ago Josef De La Vega identified a behavioral pattern in his advice to individual investors.

Joseph De La Vega 2 50pct Shrink

“As there are so many people who cannot wait to follow the prevailing trend of opinion, I am not surprised that a small group becomes an army. [Most people] think only of doing what the others do and of following their examples. . . .”
— Josef de la Vega, 1688, “Confusion de Confusiones” (the first book describing a stock market: the Amsterdam Exchange)

De La Vega's observation applies both to investor behavior and the stock price trends that their opinions drive.

About 6 years ago researchers at Empirical Research Partners found that media sentiment was correlated with future stock price drift systematically, across thousands of stocks. This makes sense intuitively - stocks with good news trend higher, and bad news stocks drift lower. The effect wasn't large (about 2.5% annually) but as we dug into it, we found it to be a rich vein - occurring in virtually every country, at most time horizons, and with larger returns for sentiment subsets. The phenomenon is called "sentiment momentum", and it is the driving force of market trends.

To see whether sentiment momentum is still apparent in markets, we updated Empirical's research last week. We found that sentiment momentum is still intact and as influential as ever (see image below). 

Russell3000 V4 Pricedrift Decile Spread Anthony Luciani 2006 2024 25pct ShrinkMethodology: For companies in the Russell 3000 sentiment was derived using natural language processing of millions of corporate references from thousands of news and social media sources in 13 languages. We averaged the sentiment over the past 30 days for each company and then ranked the companies and binned them into deciles (10% groups). The model tracked the performance of each decile over the next 90 days and averaged returns over the period 2020 - 2024 (Sep), leading to the equity curves above. We see a price drift - negative media leads prices lower, and positive media leads prices higher for individual stocks. No transaction costs were included in the simulation.

We see “media sentiment momentum” at several horizons - daily, weekly, monthly, quarterly, and yearly. In this whitepaper on the effect we go into much more detail. Notably, the effect is strongest for negative news and in India.

Sentiment momentum is one of the foundations of the 30-day predictive model that MarketPsych distributes through the London Stock Exchange Group (LSEG) - it is called the StarMine MarketPsych Media Sentiment Model. The equity curve derived from the spread between the top and bottom-ranked stock deciles of that model is plotted below.

2024 09 23 Mms Cumlative Growth Us Spread Shades 50pct Shrink

Note the steady increase in value represented by the blue line (no transaction costs included). On average the top-ranked stocks outperform the bottom-ranked. The red dotted line designates when this model was moved to production and commercial sale (started Jan 2020).

Based on the above, the news trend is the investor's friend.  Academics hypothesize that the price trends are due to investors "underreacting" to good (and bad) news. To perform better as investors, perhaps we need to act more quickly in entering good and exiting bad-news positions?  The data isn't so clear.

From our Trader Personality Tests, taken by 20,000 people online, we found that one of the best predictors of trading success was agreement with the statement "I react quickly to new ideas." We also see from behavioral portfolio analytics that portfolio mangers tend to enter high conviction positions too slowly.  They lose about 200 bps after the idea is created to fully enter the position.

So yes, the news trend is usually your friend, and it's important to move quickly and at scale (don't wait for price confirmation).  But when the news contradicts the reasons you entered the position, it's important to move quickly and exit rapidly.

We'll post some interesting insights from behavioral portfolio management software in a future newsletter.

Happy Investing!
The MarketPsych Team
support@marketpsych.com