This dataset contains the text from Federal Reserve FOMC (Federal Open Market Committee) meeting minutes and statements, collected by scraping the Federal Reserve's website. The data spans a specific period of time, providing insights into the central bank's monetary policy decisions and discussions.

Content

The dataset consists of the following columns:

• Date: The date of the FOMC meeting or statement release in the format YYYYMMDD.
• Type: Indicator for the type of document. 0 represents a statement, while 1 represents meeting minutes.
• Text: The text content of each paragraph in the meeting minutes or statements.

Acknowledgements

The data is collected from the official Federal Reserve website (https://www.federalreserve.gov) using a custom Python scraper built with BeautifulSoup.

Inspiration

This dataset can be used for various purposes, such as:

1. Analyzing the sentiment and tone of FOMC meeting minutes and statements over time.
2. Identifying key phrases and words that indicate changes in monetary policy.
3. Developing natural language processing models to predict future policy decisions based on historical data.
4. Investigating the relationship between FOMC meeting minutes/statements and financial market reactions.

Label Interpretation

LABEL_2: NeutralLABEL_1: HawkishLABEL_0: Dovish

  Citation and Contact Information





  Cite

Please cite our paper if you use any code, data, or models. @inproceedings{shah-etal-2023-trillion, title = "Trillion Dollar Words: A New Financial Dataset, Task {&} Market Analysis", author = "Shah, Agam and Paturi, Suvan and Chava, Sudheer", booktitle = "Proceedings of the 61st Annual Meeting of the Association for… See the full description on the dataset page: https://huggingface.co/datasets/gtfintechlab/fomc_communication.

Dataset Summary

For dataset summary, please refer to https://huggingface.co/datasets/gtfintechlab/federal_reserve_system

  Additional Information

This dataset is annotated across three different tasks: Stance Detection, Temporal Classification, and Uncertainty Estimation. The tasks have four, two, and two unique labels, respectively. This dataset contains 1,000 sentences taken from the meeting minutes of the Federal Reserve System.

  Label Interpretation… See the full description on the dataset page: https://huggingface.co/datasets/gtfintechlab/federal_reserve_system.

Report on operations of the Board during the year. Provides minutes of Federal Open Market Committee meetings, financial statements of the Board and combined financial statements of the Reserve Banks, financial statements for Federal Reserve priced services, information on other services provided by the Reserve Banks, directories of Federal Reserve officials and advisory committees, statistical tables, and maps showing the System's District and Branch boundaries. Also known as Policy Action Summaries.

The benchmark interest rate in Japan was last recorded at 0.50 percent. This dataset provides - Japan Interest Rate - actual values, historical data, forecast, chart, statistics, economic calendar and news.

The benchmark interest rate in India was last recorded at 5.50 percent. This dataset provides - India Interest Rate - actual values, historical data, forecast, chart, statistics, economic calendar and news.

The benchmark interest rate in Australia was last recorded at 3.60 percent. This dataset provides - Australia Interest Rate - actual values, historical data, forecast, chart, statistics, economic calendar and news.