Menu

Tracking companies through acquisitions and renames in a panel dataset

Reading 0:00
15s threshold

Putting together a panel of companies over several years, with headcount and hiring activity as the main variables. One issue I've run into is that company identities aren't always stable. Firms get acquired, merge, or rebrand, and the same entity can show up under a different name or even a new domain.

For identity resolution, I've been cross-checking Wikidata and OpenCorporates, along with SEC filings for public companies. For headcount, I'm using Coresignal's historical data, which gives me a time series for each company rather than just a single snapshot. Pulling it takes one HTTP request per company, so getting it into a Python script was pretty straightforward.

The harder question is methodological. When a company gets acquired, should I follow the acquired entity, switch to the parent company, or drop it from the dataset from that point on?…

Continue reading — create a free account

Join HashtagPLUS to read full articles, follow hashtags, vote, and join the conversation.