Zohrehvand, Amirhossein; (2020) M&As and CEOs: Machine Learning Aided Analyses of Social Media. Doctoral thesis (Ph.D), UCL (University College London).

Preview

Text ZOHREHVAND_Miors_thesis.pdf Download (2MB) | Preview

Abstract

My thesis lies at the intersection of strategy, machine learning, and social media: I examine how machine learning and social media are changing organizations and opening up new methodological avenues for strategy research. In three papers, I use Twitter data and machine learning to make theoretical and methodological contributions to strategy research. In the first paper, I use a novel synthetic control method that relies on machine learning to extend methods for analyzing M&A outcomes beyond the literature’s current focus on shareholder returns. I use more than 52 million tweets and account- ing data to illustrate applications of the method in analyzing two customer-related outcomes, i.e., customer sentiment and sales. In the other two papers, I focus on how CEOs’ interactions on social media influence their behavioral patterns and, subsequently, their strategic decisions. Social media enable executives to reach a broad audience and receive a novel form of unmediated real-time feedback from the public. In the second paper, I use a state-of- the-art (as of spring 2020) natural language processing technique, i.e., Bidirectional Encoder Representations from Transformers (BERT), to understand how feedback influences a CEOs’ communication patterns. In the third paper, I discuss that a CEO’s social media interactions influence her priorities, confidence, and attention, changing her M&A decisions.

Download activity - last month

Export as XMLCSVJSON

Download activity - last 12 months

Export as CSVXMLJSON

Downloads by country - last 12 months

Export as XMLCSVJSON

Archive Staff Only

View Item