LSEG MarketPsych Analytics

Rich sentiment data tracking news and social media in real time

brochure-image
MarketPsych
Analytics from LSEG
Rich sentiment data tracking news and social media in real-time
MarketPsych Analytics (LMA) is MarketPsych's flagship product, created for clients looking for a simplified data feed ready for quantitative analysis. First released in 2012, LMA has been revised roughly every 3 years, incorporating updates in NLP technology and expanding coverage.

Version 5 is slated for release in early 2025.
Take the market’s pulse
Buzz and sentiment scores on thousands of news and social media sources.
Easy to use
Time-series data with real-time updates, using stable entity identifiers.
Granular
Transforms unstructured text into dozens of scores – delivered every 60 seconds, hourly and daily.
Versatile use cases
Alternative data proven to drive profitable research and decision making.

At a Glance

Broad content and cutting-edge NLP
Using a patented natural language processing engine, LSEG MarketPsych Analytics turns unstructured text from thousands of curated news and social media sources into structured scores.
The data provides highly dimensional analysis of high-quality news outlets, including the full Reuters real-time markets coverage. Data coverage includes thousands of social channels and tens of millions of authors. Refined spam detection and deduplication filters based on years of hands-on experience increase score relevance and reliability.
Our feature-rich data stream tracks over a hundred sentiment scores for indices, stocks, commodities and currencies. Updated once a minute, the data can easily be used in dashboards and statistical tools for rapid investment-idea development.
New financial insights derived from the data have featured in 100+ academic papers and are available in our book “Trading on Sentiment” (Wiley Finance, 2016), also available in Chinese and Japanese.
By the numbers
Wide coverage:
– 252 countries
– 100,000+ global companies sectors and ETFs
– 62 stock indexes and sovereign bonds
– 44 currencies
– 53 commodities
– 500+ cryptocurrencies
Long history: 1998 to the present
Multi-language: 12 languages
AI-based: machine learning NLP system
Highly granular: 100+ scores
Real time: 60 second, hourly and daily updates
Sources: +4,000 news and social media outlets

Dataset Structure

For each asset tracked, MarketPsych Analytics monitors Mentions (number of references) and Sentiment (positive or negative tones). The data also tracks a range of emotional indicators – including Optimism, Fear, Uncertainty – and fundamental themes – such as earnings forecast, fundamental sentiment and interest ratesforecast – and many others.
LMA contains multiple scores for each asset class. Below are a few that our clients find useful.
Multidimensional sentiment and topics
Commodities
Acreage Cultivated
Agricultural Disease
Consumption Volume
New Exploration
Price Direction
Price Forecast
Production Volume
Supply vs Demand
Weather Damage
Safety Accident
Companies
Accounting Sentiment
Analyst Rating
Earnings Forecast
Innovation
Labour Dispute
Layoffs
Litigation
Management Sentiment
Product Sentiment
Price Forecast
Countries
Agricultural Stress
Consumer Sentiment
Economic Uncertainty
Government Corruption
Inflation Forecast
Social Inequality
Social Unrest
Terrorism
Trade Balance
Unemployment
Country Markets
Bond Buzz
Bond Sentiment
Bond Trust
Stock Index Sentiment
Stock Index Trust
Stock Index Uncertainty
Rates Buzz
Interest Rate Forecast
Monetary Policy Loose
Real Estate Sentiment
Currencies
Carry Trade
Currency Peg Instability
Emotion vs Fact
Future vs Past
Long vs Short
Long vs Short Forecast
Market Risk
Price Direction
Price Forecast
Price Momentum
Cryptocurrencies
Adoption
Attack
Code Sentiment
Code Upgrade
Developer Sentiment
FOMO
HODL
Regulatory Crackdown
Scam
Transaction Speed

Achieve more with data

Why MarketPsych Analytics?
MarketPsych produces innovative data feeds by performing high-speed AI-based natural language processing on millions of daily news and social media messages. Data customers hail from 25+ countries.
Our data, which extracts meaningful measures from the global information flow, underlies bespoke alpha-generation models, risk-management tools and insightful research.
Why LSEG Data & Analytics?
LSEG Data & Analytics, now part of London Stock Exchange Group, is one of the world’s largest providers of financial markets data and infrastructure. With $6.25 billion in revenue, over 40,000 customers and 400,000 end users across 190 countries, we provide information, insights and technology that enable customers to execute critical investing, trading and risk decisions with confidence.
We draw on cleaner, richer, more readily accessible data to support the use of AI and machine learning – helping our customers gain greater insight, boost performance and fuel more rapid innovation.

Popular client use cases

Sample client profiles
Global macro hedge funds
Understand market cycles based on the psychological perceptions in each country, market and asset class. Make superior global asset allocation decisions.
Forex and currency
Identify news and social media flow that impacts investment theses. Find arbitrage opportunities by monitoring macroeconomic imbalances and attitudes towards countries and currencies.
Commodity and energy firms
Track global information flow about agricultural and energy products, including supply and demand issues, industrial accidents and weather damage, better-time trend emergence, exhaustion and reversals.
Cryptocurrency investors and exchanges
Deploy artificial intelligence and machine learning strategies to anticipate swings in sentiment. Hedge as investor enthusiasm wanes. Identify trending coins and innovative technologies.
Data experts using our products
Quant traders/investors
Better identify changing market cycles, augment alpha generation strategies and improve risk models. Refine momentum, mean reversion and volatility strategies.
Research analysts
Find insights in the crowd and see what is being missed. Monitor market-moving events. Identify the specific price drivers of an asset.
Risk managers
Identify the emergence of new risk and dangerous events. Monitor speculative and crowding risk across assets in which your firm is positioned.
Economists
Monitor risk perceptions across global economies, sovereign bonds, and currencies. Identify changes in key macroeconomic themes such as global interest rate forecasts and unemployment. Update leading economic indicators with news and information as it happens (nowcasting).

From data to outcomes

Find outperforming assets
Financial markets take time to fully digest news. This is true for both good news and bad – and presents a unique opportunity for those with a keen eye for sentiment analysis.
A practical example is shown in the figure to the right – depicting a study of 30,000 global companies over the 2006-2024 period.
MarketPsych calculated each company’s average sentiment over the previous month. Using this score, they grouped companies into deciles, and we averaged each group’s subsequent 90-day stock price performance relative to the groups’ average.
Stocks with the most positive media sentiment over the prior month significantly outperformed the most negative over the following 90 days.
The price spread is just one actionable opportunity among many – made possible by extracting signal from the noise in unstructured news data. Buy-side and research firms can develop their own sentiment-based signals using such insights.
Figure: Average subsequent price drift of stocks ranked by 30-day sentiment (news_social) average

From data to outcomes

Time trends with sentiment
Media sentiment sometimes leads prices at major turning points. Sentiment moving averages can help identify tops and bottoms, improving asset allocation and risk-on/risk-off timing.
For example, the graph to the right shows the performance of the S&P 500 (orange daily bars) versus media sentiment during 2020.
The smooth lines show two moving averages of media sentiment (30 and 90 days).
Green shading appears between the averages when the 30-day average is above the 90-day (meaning that recent sentiment is rising), and red shading appears when below (recent sentiment is falling).
Note that media sentiment generated a risk-off signal, declining steeply before the S&P 500 stock market’s sell-off in response to Covid-19’s spread.
Figure: United States price (.SPX) vs United States stock index sentiment (news_social) 30/90 days

From data to outcomes

Screen for opportunities
Visualisation tools and screeners help investors unearth opportunities and risks. For example, the Trust indicator heat map to the right shows the percentage of all language conveying trust in U.S. companies during Q1 2021. Firms whose media coverage scores highly on trust are shown in bright green, while those with lower trust scores appear in brighter red.
The map was produced during the Covid-19 pandemic and the U.S. presidential transition. Note that vaccine makers and biotech firms ranked highly on trust, while information platforms including Twitter and Facebook ranked low.
Data analysts can monitor dozens of business and supply chain risks in real-time using such visualisation tools and screeners.
Figure: Global company media-based trust, Q1 2021. Bright green denotes high trust, bright red denotes low trust.