Title: Analyse ECB's official press statements
Author: Hans-Peter Hoellwirth
Date: 06.2017
Web scrapping official press statements published at https://www.ecb.europa.eu/press/pressconf/2017/html/index.en.html for the years 1998-2017.
Latent Dirichlet Allocation (LDA) and Structural Topic Model (STM) for 3 and 10 topics
3 topics: Topics are mainly associated to presidencies
10 topics: Most topics are associated to major events and policies in ECB's history
Run in the following order:
- 01_web_scraping.py: Creates file data/combined.csv (optional if already existing)
- 02_pre_processing.py: Requires file data/combined.csv to exist
- 03_lda.py: Takes output of previous step (text_process) as input
- 04_stm.R: Requires file data/combined.csv to exist. Independently executable from steps 2 and 3
Note: Before execution, update the working directory in each code file.