Datasets
All datasets link to verifiable public sources. No data is fabricated. Each dataset includes a full data dictionary, methodology notes, and example code.
15
Datasets
5
Domains
53,506
Downloads
COVID-19 Outbreak Dataset
Daily confirmed cases and deaths for 289 countries and territories from January 2020 to March 2023. Source: JHU CSSE COVID-19 Data Repository, the definitive public epidemiological dataset of the pandemic.
NBA Player Efficiency & Advanced Stats (2001–2023)
Season-level advanced statistics for all NBA players with qualifying minutes played. Includes PER, True Shooting %, Win Shares, Box Plus/Minus, and VORP. Sourced from the Basketball Reference advanced stats table, covering 2001 through 2023.
Hudson's Bay Company Fur Returns (1845–1935)
Annual pelt counts for snowshoe hare and Canadian lynx from Hudson's Bay Company trading posts, 1845–1935. The canonical empirical dataset for predator-prey oscillation dynamics and Lotka-Volterra model validation.
NBA Shot Chart Dataset (2004–2023)
Every tracked shot attempt in NBA regular season games from 2004 to 2023: location (x,y coordinates), shot type, result, shooter, zone, and game context. 20 seasons, approximately 4 million shot attempts.
Sector Rotation & Returns Dataset
Monthly total returns for all 11 GICS equity sectors from January 2000 to December 2023, with momentum signals, volatility estimates, and fund flow proxies. Sourced from FRED sector indices and ETF net-asset-value changes.
NBA Player Tracking Aggregate Statistics
Aggregated player tracking statistics from the NBA Stats API, including speed, distance, defensive matchup quality, and spatial metrics. Available from 2013–14 through 2023–24.
Good Judgment Project Forecasting Dataset
Aggregated probabilistic forecast data from the Good Judgment Project (GJP), a DARPA-sponsored geopolitical forecasting tournament. Includes resolution rates, Brier scores, and calibration statistics for ~500 questions across 4 years.
Market Reaction to Macro Announcements
Intraday S&P 500 price reactions in the 30-minute window around 14 categories of scheduled macroeconomic announcements, 2010–2023. Includes CPI, FOMC, NFP, GDP, PPI, and 9 additional announcement types.
Economic Forecast Accuracy Dataset
Compilation of economic forecasts and their outcomes for US GDP growth, inflation, and unemployment from major forecasting institutions (Fed, IMF, CBO, Blue Chip consensus), 1995–2023.
Earnings Call Sentiment Dataset
Quarterly earnings call transcripts for S&P 500 companies, scored using the Loughran-McDonald financial sentiment dictionary. Covers Q1 2018 through Q4 2023, providing tone, uncertainty, litigious, and complexity metrics alongside sector classification.
Multi-Instrument Spectrogram Dataset
Precomputed spectrograms and acoustic feature vectors for 12 orchestral instruments, each at multiple pitches and dynamic levels. Derived from the RWC Musical Instrument Sound database.
Multi-Armed Bandit Decision Dataset
Behavioral decision data from a four-armed bandit fMRI experiment with 14 participants facing explore-exploit tradeoffs over 300 trials each. Published supplementary data from Daw et al. (2006, Nature Neuroscience).
Piano & Keyboard Harmonic Analysis Dataset
Spectral analysis of piano and keyboard recordings across 88 keys, with harmonic series decomposition, inharmonicity coefficients, and overtone amplitude envelopes. Derived from University of Iowa Electronic Music Studios recordings.
Experimental Tumor Growth Dataset
Digitized tumor volume time series from 62 experimental studies, covering breast, lung, colon, and melanoma xenografts in mouse models. Source: Benzekry et al. (2014) supplementary data (PLOS Computational Biology).
Hindustani Raga Frequency Dataset
Chroma, MFCC, and spectral features extracted from verified Hindustani classical music recordings across 10 ragas. Sourced from the Compmusic Dunya dataset with expert annotations.