Capturing international influences in U.S. monetary policy through a Natural Language Processing approach

Published: 26 August 2026| Version 1 | DOI: 10.17632/j7m28r3j4d.1
Contributors:
Laurent Ferrara, Nicolas de Roux

Description

Replication Files for ------------------------------------- -- Capturing international influences in U.S. monetary policy through a Natural Language Processing approach -- ------------------------------------- Laurent Ferrara and Nicolas de Roux ------------------------------ Contents and Replication Instructions This repository contains the replication code and dataset for the study. The current version reflects the code and data as of 2026-08-25. This replication package includes: (i) a detailed ReadMe file with step-by-step instructions for replication; (ii) a mapping of scripts to outputs, specifying which script generates each figure and table in the paper and its appendices; (iii) all datasets and source code needed to reproduce the analysis end to end, from the FOMC text corpora through sentence classification to the Taylor-rule estimations; and (iv) a test suite that checks every regenerated table cell by cell against the numbers the manuscript prints. The pipeline runs in five stages. Because the intermediate outputs are shipped - the processed corpora, the labelled sentences, the fitted classifiers and topic model, and the attention indices - the usual entry point is the last one: "uv run run_all.py --stage E" reproduces every table and figure in about five minutes. Earlier stages can be re-run individually; the ReadMe documents what each needs and how long it takes. ------------------------------ Code Annotations and System Requirements The code is extensively commented and needs no high-performance computing. The Python environment installs from a lock file with exact versions ("uv sync"); only the Taylor-rule script additionally requires R, reached through rpy2 and documented in the ReadMe. Replication runs on a standard laptop: the final stage takes about five minutes, and re-deriving everything from the raw text several hours, dominated by transcript parsing and classifier training. Shortcuts are provided so that no step is a prerequisite for those after it - cached feature matrices (1.8 MB) rebuild the classifier comparison without loading the 430 MB pretrained embedding, and cached GPT-3.5 labels remove any need for an API key. Two inputs are not redistributed. MSCI World index levels obtained through Bloomberg cannot be shared under the licence terms; a format stub and full retrieval instructions are included, and the three affected exhibits are identified in the ReadMe. The FOMC transcript PDFs (about 922 MB) are omitted for size only - the extracted text that the code actually consumes is included, and the ReadMe gives the command to re-download the PDFs.

Files

Steps to reproduce

See the contained README.md for a complete overview. Method in brief: 1. Install uv and the Python environment. All package versions are pinned in uv.lock: curl -LsSf https://astral.sh/uv/install.sh | sh uv sync 2. Install R (needed only for the Taylor-rule tables; every other script runs without it). Only base R is used, so there are no R packages to install. sudo apt-get install r-base r-base-dev # Debian / Ubuntu brew install r # macOS Then check the setup, which reports the detected R version and any missing input: uv run run_all.py --check-env 3. Unzip the archive into a folder of your choice. It contains: -- README.md, DATA_SOURCES.md: instructions, and the provenance of every input. -- interfed/: the library (text processing, loaders, Taylor specifications, rendering). -- scripts/: numbered entry points, one per pipeline step, from 00_scrape.py to 07_export_panel.py. Each takes --help and --list. -- Data/: all inputs - FOMC minutes and transcript text, 5,000 hand-labelled sentences, cached GPT-3.5 labels, macroeconomic series, and third-party data with its citations and terms. -- Models/: the fitted classifiers and topic model behind the published results, plus cached feature matrices. -- Output/: tables/, plots/, panel/ and artifacts/, all regenerated by the scripts. -- tests/: the verification suite. 4. Reproduce every table and figure (about five minutes): uv run run_all.py --stage E Individual exhibits can be produced on their own, for example: uv run scripts/04_taylor_tables.py --table taylorresults README.md maps each figure and table in the paper, appendix and online appendix to the exact command that generates it. 5. Verify the output against the published numbers: uv run pytest tests -q

Institutions

Categories

Economics, International Economics, Monetary Economics

Licence