fabhaus/equities_5m_stockprices

该数据集名为Equities 5-Minute Stock Prices (Enriched)，是一个美国股票（普通股和ETF）的日内5分钟OHLCV（开盘价、最高价、最低价、收盘价、成交量）条形数据集合，每行数据还融合了技术指标、宏观经济序列、SEC公司基本面数据以及保守的内在价值估计。数据集覆盖时间窗口为2024年1月至2026年3月（共27个月份分片），包含约13.3万个股票代码，总行数约2.787亿行。数据以每月一个JSONL分片的形式组织，每行对应一个股票在一个5分钟时间条的数据。列包括：标识符（如symbol、datetime）、OHLCV数据、技术指标（如移动平均线、RSI、MACD等）、宏观经济指标（如CPI、失业率、国债收益率等）、公司基本面（如收入、净利润、现金流等）以及公司价值估计（如内在价值、估值差距等）。数据集适用于时间序列预测和表格回归任务，主要用于金融研究、特征工程等场景，但需注意数据是as-of连接（无前瞻性）、指标有预热期、未进行公司行动调整、部分股票缺少基本面数据等限制。数据来源包括多个历史条形API、BLS、美国财政部和SEC EDGAR，使用时需遵守相关许可。The dataset is named Equities 5-Minute Stock Prices (Enriched), which provides intraday 5-minute OHLCV (Open, High, Low, Close, Volume) bars for U.S. equities (common stocks and ETFs), enriched per row with technical indicators, macroeconomic series, SEC fundamentals, and a conservative intrinsic-value estimate. It covers the period from January 2024 to March 2026 (27 monthly shards), includes approximately 13.3k tickers, and contains about 278.7 million rows. Data is organized as one JSONL shard per calendar month, with each row representing a single stock at one 5-minute bar. Columns include: identifiers (e.g., symbol, datetime), OHLCV data, technical indicators (e.g., moving averages, RSI, MACD), macroeconomic indicators (e.g., CPI, unemployment rate, Treasury yields), company fundamentals (e.g., revenue, net income, cash flow), and company value estimates (e.g., intrinsic value, valuation gap). The dataset is suitable for time-series forecasting and tabular regression tasks, primarily intended for financial research and feature engineering. Caveats include as-of joins (no look-ahead), indicator warm-up periods, lack of corporate-action adjustments, missing fundamentals for some tickers, and all timestamps in UTC. Sources include historical bar APIs, BLS, U.S. Treasury, and SEC EDGAR, with licensing terms from each upstream source.