README.md

metadata

dataset_info:
  features:
    - name: date
      dtype: date32
    - name: stock
      dtype: string
    - name: title
      dtype: string
    - name: content
      dtype: string
    - name: trading_date
      dtype: timestamp[ns]
    - name: exact_trading_date
      dtype: timestamp[ns]
    - name: ts_-30
      dtype: float64
    - name: ts_-29
      dtype: float64
    - name: ts_-28
      dtype: float64
    - name: ts_-27
      dtype: float64
    - name: ts_-26
      dtype: float64
    - name: ts_-25
      dtype: float64
    - name: ts_-24
      dtype: float64
    - name: ts_-23
      dtype: float64
    - name: ts_-22
      dtype: float64
    - name: ts_-21
      dtype: float64
    - name: ts_-20
      dtype: float64
    - name: ts_-19
      dtype: float64
    - name: ts_-18
      dtype: float64
    - name: ts_-17
      dtype: float64
    - name: ts_-16
      dtype: float64
    - name: ts_-15
      dtype: float64
    - name: ts_-14
      dtype: float64
    - name: ts_-13
      dtype: float64
    - name: ts_-12
      dtype: float64
    - name: ts_-11
      dtype: float64
    - name: ts_-10
      dtype: float64
    - name: ts_-9
      dtype: float64
    - name: ts_-8
      dtype: float64
    - name: ts_-7
      dtype: float64
    - name: ts_-6
      dtype: float64
    - name: ts_-5
      dtype: float64
    - name: ts_-4
      dtype: float64
    - name: ts_-3
      dtype: float64
    - name: ts_-2
      dtype: float64
    - name: ts_-1
      dtype: float64
    - name: ts_0
      dtype: float64
    - name: ts_1
      dtype: float64
    - name: ts_2
      dtype: float64
    - name: ts_3
      dtype: float64
    - name: ts_4
      dtype: float64
    - name: ts_5
      dtype: float64
    - name: ts_6
      dtype: float64
    - name: ts_7
      dtype: float64
    - name: ts_8
      dtype: float64
    - name: ts_9
      dtype: float64
    - name: ts_10
      dtype: float64
    - name: ts_11
      dtype: float64
    - name: ts_12
      dtype: float64
    - name: ts_13
      dtype: float64
    - name: ts_14
      dtype: float64
    - name: ts_15
      dtype: float64
  splits:
    - name: train
      num_bytes: 101756779
      num_examples: 82895
    - name: test
      num_bytes: 55328208
      num_examples: 22645
  download_size: 91629781
  dataset_size: 157084987
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: test
        path: data/test-*

US Stock News With Price

A stock news and price dataset design for the alignment between financial news and the market. Please only use for academic purpose.

1. Data Description

• date: The date of the news published.
• stock: The symbol of the stocks the news related to. (checked by whether title or content has the company information.
• title: The title of the news.
• content: The content of the news.
• trading_date: Here is the assumed trading date, which should be the next date of the publish date.
• exact_trading_date: The exact next trading date after the news was made public.
• ts_{-30...-1}: Stock prices before the exact trading date. (30 trading days)
• ts_0: Stock prices of the exact trading date.
• ts_{1...15}: Stock prices after the exact trading date. (15trading days)

2.Data Split

• Train phrase: exact_trading_date < = '2021-12-31'.
• Test phrase: exact_trading_date > '2021-12-31'.

3. Citation

• To cite this dataset, please use the following BibTeX entry:

@misc{wang2024huggingface,
    author       = {Guoxuan Wang},
    title        = {US Stock News With Price: A stock news and price dataset design for the alignment between financial news and the market.},
    year         = {2024},
    url          = {https://huggingface.co/datasets/oliverwang15/us_stock_news_with_price},
    note         = {Available at Hugging Face.}
  }