MSc Thesis Defense: Erfan Tarmohammadi, LLM-DRIVEN EXTRACTION AND FINANCIAL APPLICATION OF MERGERS AND ACQUISITIONS EVENTS AND ACTORS FROM EDGAR, Date & Time: 14 July, 2026 – 1:30 PM, Place: L063

LLM-DRIVEN EXTRACTION AND FINANCIAL APPLICATION OF MERGERS AND ACQUISITIONS EVENTS AND ACTORS FROM EDGAR

Erfan Tarmohammadi
Data Science, MSc Thesis, 2026

Thesis Jury

     Asst. Prof. Onur VAROL (Thesis Advisor)

  Prof. Şerif Aziz ŞİMŞİR

  Assoc. Prof. Emre EKİNCİ

Date & Time: 14th July, 2026 – 1:30 PM

Place: FENS L063

Keywords : Mergers, Acquisitions, SEC Filings, EDGAR, LLMs, Private Phase

Abstract

This thesis studies the private phase of mergers and acquisitions by introducing a novel large language model–based framework to extract key negotiation events and actors from SEC disclosures and analyzes how these private-phase dynamics shape competition, pricing, and renegotiation outcomes. It addresses a gap in the literature by replacing traditional hand-collected, static measures with scalable actor- and event-level data constructed from narrative filings with LLMs. The evidence shows that bidder participation and external advisor involvement have increased over time, accompanied by more price-specific offers and price revisions. The magnitude and frequency of price revisions are systematically related to the number of bidders, indicating that competition influences price formation. Negotiations follow a structured and economically intuitive sequence rather than unfolding as a random process. The thesis contributes methodologically by demonstrating how narrative corporate disclosures can be transformed into economically meaningful structured data using LLMs and substantively by providing large-sample evidence that opens the “black box” of private-phase M\&A negotiations.