Extracting Merger Deal Initiation Data with LLMs

The merger process could be initiated by acquirers (buyer firm) or targets (selling firm). Finance scholars collect the deal-initiating party data from the SEC documents manually. However, this process is tedious and subjective, leading to inconsistencies across studies. Most importantly, hand collection of data is a major obstacle for performing research in this field. This project aims to automate this process. Team members are expected to use their coding skills to scrap relevant forms from the SEC's website, locate the relevant text from these forms, and use LLMs to extract the deal-initiating party. The performance of the algorithm will be compared with the dataset collected manually. Excellent coding skills, as well as familiarity with LLMs, are required.