cb-text-analytics

Topic modeling of central bank communications, tracking how themes and language
evolve across FOMC and RBNZ statements over time.

What This Does

Applies LDA topic modeling to identify recurring themes in central bank statements
and track how those themes shift across policy eras.

Data

• Federal Reserve FOMC statements: 2014–2017, 2023–2025
• Reserve Bank of New Zealand OCR decisions: 2006–2012

Both datasets are included under usa-central-bank/ and nz-central-bank/.

Analysis

• fomc_analysis.ipynb — Topic modeling of Fed statements
• rbnz_analysis.ipynb — Topic modeling of RBNZ statements

Methods

Built with Python. Uses LDA for topic extraction and time-series analysis
to track topic prevalence across statement dates.

Status

Active research project. FOMC analysis complete, RBNZ analysis in progress.

License

CC0 — data and code are public domain.