Use of modern machine learning techniques to predict the occurrence and outcome of corporate takeover events
Name
1256665200-MIT.pdf
Size
488.84 KB
Format
Adobe PDF
Checksum (MD5)
56a195210fcaa9522975382420e7f811
Author(s)
Geha, Georges.
Advisor(s)
David Jean Joseph Thesmar.
Date Issued
2021
Publisher
Massachusetts Institute of Technology
Abstract
The objective of this project is to use machine learning to predict the occurrence of corporate takeovers. The findings show that random forest yields the best predictions out-of-sample based on the area under the curve (AUC) metric. As such, 8 independent variables are considered statistically significant. A time series machine learning approach is also used at the end of the study to predict these events in 2019 based on each company's data from 2010 to 2018. Random forest is still determined as the model with the best out-of-sample performance. A strategy of investing equal amounts across the companies predicted to be takeover targets in 2019 based on the model yields a profit of 7.4%.
Description
Thesis: M. Fin., Massachusetts Institute of Technology, Sloan School of Management, Master of Finance Program, February, 2021
Cataloged from the official PDF version of thesis.
Includes bibliographical references (page 23).
Subjects
Sloan School of Management. Master of Finance Program.
MIT Department
Sloan School of Management
Terms of Use
MIT theses may be protected by copyright. Please reuse MIT thesis content according to the MIT Libraries Permissions Policy, which is available through the URL provided.
Persistent DSpace Link
