Calomiris, Çakir Melek, and Mamaysky, 2022, "Big Data Meets the Turbulent Oil Market"

Topic model from the 2022 version of the energy paper.  The topic number-label mapping:

1. Company (Co)

2. Global Oil Markets (Gom)

3. Environment (Env)

4. Energy/Power Generation (Epg)

5. Crude Oil Physical (Bbl)

6. Refining & Petrochemicals (Rpc)

7. Exploration & Production (Ep)

The topic model in the 2025 version of the paper (available here) has changed somewhat, but the list of energy words is the same.

File with data (markets, physical oil, text): transformed_data_prices.csv (link temporarily removed)