Research Preparation Exam: Gene Moo Lee, GDC 5.516

Contact Name: 
Lydia Griffith
Date: 
Dec 4, 2013 1:00pm - 3:00pm

Research Preparation Exam: Gene Moo Lee

Title: Towards Modeling M&A in High Tech Industry
Date & time: 1 PM, December 4th, 2013
Place: GDC 5.516
Committee members: Andrew B. Whinston (chair), Lili Qiu, Vladimir Lifschitz

Abstract:

Mergers and acquisitions (M&A) play an important role in high tech
industry. Early stage ventures successfully exit by M&A. In the
meantime, established companies pursue innovation by acquiring
promising startups. This trend calls for a better understanding of the
matching between sellers and buyers in M&A transactions. In this talk,
we build a model of M&A transactions based on novel proximity measures
and validate the model with a large dataset of company profiles and
M&A deals in high tech industry. Our contribution is threefold.

First, we propose a novel business proximity, which measures the
closeness between high tech firms. In doing so, we apply topic
modeling algorithm on business descriptions. We also construct three
additional proximity metrics based on geographic closeness, social
linkages, and investment relations. Empirically, we analyze the the
explanatory power of each measure to  M&A transactions.

Second, we develop a statistical model to understand the matching
problem in M&A transactions based on the proximity measures. The
relational nature of M&A transactions invalidates the independence
assumption underlying the conventional binary response econometric
models such as probit or logit regression. To overcome this issue, we
employ exponential random graph models (ERGM) to facilitate our
empirical analysis. To best of our knowledge, this is the first work
to apply ERGM in M&A literature.

Lastly, we propose a two-sided platform for M&A matching. To increase
the market efficiency, we develop an interface VentureMap to let
buyers and sellers search their potential partners. As a future
direction, we will extend our model and proximity measures to conduct
link prediction on M&A networks.