CRYPTO
This unique dataset tracks and quantifies the impact of social media on crypto assets and generates ALPHA providing a detailed analysis of activities of influencers, financial professionals, retail investors, and bots across Social Media platforms.
Datasets -> Crypto
Introduction
The dataset includes sentiment scores derived from social media content related to cryptocurrencies. The output provides time series sentiment data, which allows users to track sentiment over time, including historical trends. This aspect is crucial for analyzing how sentiment correlates with market movements. ZENPULSAR's methodology includes identifying content from influencers, financial professionals, retail investors, and bots. The output might differentiate these sources, providing insights into how different types of social media participants influence sentiment.
Data Summary
Start Date: March 2009
Asset Coverage: 80 Cryptocurrencies
Delivery Method: API or BIGQUERY
Resolution: 15 Minutes
Source List
The dataset includes sentiment scores derived from social media content related to cryptocurrencies. The output provides time series sentiment data, which allows users to track sentiment over time, including historical trends. This aspect is crucial for analyzing how sentiment correlates with market movements. ZENPULSAR's methodology includes identifying content from influencers, financial professionals, retail investors, and bots. The output might differentiate these sources, providing insights into how different types of social media participants influence sentiment.
Data Summary
Start Date: March 2009
Asset Coverage: 80 Cryptocurrencies
Delivery Method: API or BIGQUERY
Resolution: 15 Minutes
Source List
- Ptt.cc
- Seeking Alpha
- Telegram
- Youtube
- Sina Weibo
- Guba.Eastmoney
- JRJ.com
- Yicai.com (China Business Network)
- Shanghai Securities Journal
- ETNet
- Finet HK
- Finance.sina
- Mingpao.com
Sentiment Quintile Crypto Portfolio Strategy
Strategy concept and implementation details
The core concept of this strategy is to identify an approach that could be appealing for testing by our typical clientele: medium-sized crypto funds that employ quantitative methods alongside fundamental-based strategies. Therefore, the strategy is designed to embody straightforward sentiment-based logic without being prone to over-optimization, ensuring it is accessible and executable by such clients.
This strategy capitalizes on the sentiment ratio (bullish vs. bearish) of various crypto assets. It involves ranking the average sentiment over the four hours preceding portfolio rebalancing. Assets in the top quintile are allocated long positions, while those in the bottom quintile are assigned short positions. Within each quintile, position sizes are distributed equally. Portfolio rebalancing occurs every 8 hours.
The assets involved in this strategy include 15 of the most liquid and popular coins:
'ADAUSDT', 'ALGOUSDT', 'BNBUSDT', 'BTCUSDT', 'DOGEUSDT', 'DOTUSDT', 'EOSUSDT', 'ETHUSDT', 'LINKUSDT', 'LTCUSDT', 'MINAUSDT', 'SOLUSDT', 'THETAUSDT', 'XLMUSDT', 'XRPUSDT'.
Regarding commissions, the strategy adheres to the standard trading fees on Binance for spot trading, which is typically 0.1%. However, by utilizing BNB to cover trading fees, users benefit from a 25% discount, reducing the trading fee to 0.075% per transaction.
The portfolio is initialized with a size of $1,000,000 USD. The trading period extends from January 1, 2021, to January 1, 2024, utilizing hourly market data from Binance. The strategy was modeled and backtested locally using the Lean framework by QuantConnect.
Strategy concept and implementation details
The core concept of this strategy is to identify an approach that could be appealing for testing by our typical clientele: medium-sized crypto funds that employ quantitative methods alongside fundamental-based strategies. Therefore, the strategy is designed to embody straightforward sentiment-based logic without being prone to over-optimization, ensuring it is accessible and executable by such clients.
This strategy capitalizes on the sentiment ratio (bullish vs. bearish) of various crypto assets. It involves ranking the average sentiment over the four hours preceding portfolio rebalancing. Assets in the top quintile are allocated long positions, while those in the bottom quintile are assigned short positions. Within each quintile, position sizes are distributed equally. Portfolio rebalancing occurs every 8 hours.
The assets involved in this strategy include 15 of the most liquid and popular coins:
'ADAUSDT', 'ALGOUSDT', 'BNBUSDT', 'BTCUSDT', 'DOGEUSDT', 'DOTUSDT', 'EOSUSDT', 'ETHUSDT', 'LINKUSDT', 'LTCUSDT', 'MINAUSDT', 'SOLUSDT', 'THETAUSDT', 'XLMUSDT', 'XRPUSDT'.
Regarding commissions, the strategy adheres to the standard trading fees on Binance for spot trading, which is typically 0.1%. However, by utilizing BNB to cover trading fees, users benefit from a 25% discount, reducing the trading fee to 0.075% per transaction.
The portfolio is initialized with a size of $1,000,000 USD. The trading period extends from January 1, 2021, to January 1, 2024, utilizing hourly market data from Binance. The strategy was modeled and backtested locally using the Lean framework by QuantConnect.
Pulse Sentiment
This dataset offers sentiment scores from the complete asset universe-related social media, enabling tracking of sentiment over time and analysis of its correlation with market movements, with insights into how influencers, financial professionals, investors, and bots impact sentiment. Request ->
Dominating Sentiment
This dataset analyzes sentiment from social media, tracking trends over time and correlations with market movements. Metrics like Dominating sentiment provide nuanced views of overall sentiment direction, considering bullish, bearish, and neutral mentions. Request ->
Influencers
This dataset analyzes the impact of key opinion-makers on retail investors and sentiment. It gathers data from over 10 million social media accounts, offering insights into topics covered by influencers, including topic popularity within specific influencer groups. Request ->
News and Reactions
This dataset focuses on trending financial articles among influencers and professional investors, capturing social media momentum and providing early market signals. With a 99% data accuracy rate and specialized financial asset focus, it offers comprehensive insights through news sentiment and social media reactions. The dataset includes news text, source, sentiment, and labels like virality and popularity among professionals. Request ->
Oracle
This dataset provides detailed sentiment analysis and classifying mentions into facts, opinions, or predictions. It offers comprehensive insights including sentiment scores (neutral, bullish, bearish), asset codes, and fact_level labels indicating mention nature (fact, opinion, prediction) from different data sources. Request ->
Labelled Posts
This dataset includes post text and AI-generated labels indicating sentiment (bullish, bearish, neutral), authenticity (bot or non-bot), fact level (opinion, prediction, fact), and spam classification. Request ->
This dataset offers sentiment scores from the complete asset universe-related social media, enabling tracking of sentiment over time and analysis of its correlation with market movements, with insights into how influencers, financial professionals, investors, and bots impact sentiment. Request ->
Dominating Sentiment
This dataset analyzes sentiment from social media, tracking trends over time and correlations with market movements. Metrics like Dominating sentiment provide nuanced views of overall sentiment direction, considering bullish, bearish, and neutral mentions. Request ->
Influencers
This dataset analyzes the impact of key opinion-makers on retail investors and sentiment. It gathers data from over 10 million social media accounts, offering insights into topics covered by influencers, including topic popularity within specific influencer groups. Request ->
News and Reactions
This dataset focuses on trending financial articles among influencers and professional investors, capturing social media momentum and providing early market signals. With a 99% data accuracy rate and specialized financial asset focus, it offers comprehensive insights through news sentiment and social media reactions. The dataset includes news text, source, sentiment, and labels like virality and popularity among professionals. Request ->
Oracle
This dataset provides detailed sentiment analysis and classifying mentions into facts, opinions, or predictions. It offers comprehensive insights including sentiment scores (neutral, bullish, bearish), asset codes, and fact_level labels indicating mention nature (fact, opinion, prediction) from different data sources. Request ->
Labelled Posts
This dataset includes post text and AI-generated labels indicating sentiment (bullish, bearish, neutral), authenticity (bot or non-bot), fact level (opinion, prediction, fact), and spam classification. Request ->
Request API Trial
Get In Touch
Request Dataset
Request Dataset