The goal of this project is to replicate research done by Narasimhan Jegadeesh (Emory University) and Di Wu (University of Michigan) [Deciphering Fedspeak: The Information Content of FOMC Meetings (2017)](https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2939937). The researchers analyze text from Federal Open Market Committee (FOMC) meeting minutes, from 1994 to 2015, by applying topic modeling, sentiment analysis, and regression modeling.

text_mining_monetary_policy.Rmd - executable Rmarkdown file with methodology, findings, and commented code.
fomc_meeting_and_release_dates.csv - FOMC meeting and minutes release schedule sourced from federalreserve.gov
fomc_minutes.zip - text files of minutes from 1994 to 2009 sourced from https://stanford.edu/~rezab/
lemmatization-en.txt - root words for stemming sourced from https://github.com/michmech/
macro_variables.zip - macro variables such as unemployment, recession indicator, etc.