Norwegian School of Economics Bergen - Are seasoned equity offerings predictable? - Predicting future SEOs with machine learning algorithms

Aug 4, 2026

This master thesis examines whether seasoned equity issuance in the United States is predictable using machine learning. Logistic regression, decision trees, random forest, and XGBoost are used to model the likelihood of future SEOs, and model performance is evaluated with AUC. The best model, XGBoost, reaches an AUC of 0.72, while random forest achieves 0.71, indicating suitability of nonlinear methods. The study also assesses stock-return impacts via linear regressions and difference-in-differences, finding no conclusive evidence.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages60

Reading Time2 h 31 min

AccessFree to download

Introduction

Literature review

Theory

Sources of funding

Seasoned equity offering (SEO)

Machine learning models

Decision trees

Random forest

XGBoost

Threshold value

Data

Data description

Feature selection

Variable explanation

Methodology

Data collection, cleaning and processing

Benchmark model

Tuning of models

Optimal threshold

Evaluation

Implications of results

Results

Norwegian School of Economics Bergen, Spring 2023 Are seasoned equity offerings predictable? Predicting future SEOs with machine learning algorithms Peder Hernholm and Andreas Ore Wormsen Supervisor: Walt Pohl Master thesis in Economic and Business Administratioin Major in Financial Economics NORWEGIAN SCHOOL OF ECONOMICS This thesis was written as a part of the Master of Science in Economics and Business Administration at NHH. Please note that neither the institution nor the examiners are responsible − through the approval of this thesis − for the theories and methods used, or results and conclusions drawn in this work. Abstract This master thesis explores the predictability of seasoned equity issuance in the United States using the machine learning methods based on logistic regression, decision-trees, random forest and XGBoost. In addition, we investigate the practical value of predicting seasoned equity offerings. Our results show a benefit from employing machine learning for this purpose, with the best performing model (XGBoost) achieving an AUC of 0.72. The random forest model demonstrated similar capabilities with an AUC of 0.71, indicating that sophisticated nonlinear models are suited for this type of prediction problem. Further, the impact of seasoned equity offerings on stock returns is analyzed to identify the possible benefits our models provide. Our efforts included two linear regressions using separate data samples, and one difference-in-differences estimation. These tests failed to provide conclusive evidence; however existing literature implies a negative effect on stock returns from seasoned equity offerings. This thesis contributes to the extensive research conducted on the topic of seasoned equity offerings. While there are no directly comparable publications, we utilize existing literature to improve our thesis and to reflect on our findings. With this thesis we facilitate and encourage further research on this relatively unexplored area of seasoned equity offerings. Acknowledgement This thesis is the final project of our Master of Science degree in financial economics at NHH. The research question was intentionally developed to combine finance and machine learning, as we both are fascinated by this subject area. Working on this thesis for the past semester has allowed us to improve on an important and quickly developing area of finance, for which we are grateful. Finally, we would like to thank our supervisor, Walt Pohl, for his quality guidance and commitment to the project. In addition, we would like to thank family and friends for supporting us in our academical endeavors. Contents 1. INTRODUCTION.......................................................................................................................................1 1.1 LITTERATURE REVIEW ..........................................................................................................................2 2. THEORY .....................................................................................................................................................4 2.1 SOURCES OF FUNDING ..........................................................................................................................4 2.2 SEASONED EQUITY OFFERING (SEO) ....................................................................................................4 2.3 MACHINE LEARNING MODELS ..............................................................................................................5 2.3.1 Decision trees .................................................................................................................................5 2.3.2 Random forest ................................................................................................................................. 7 2.3.3 XGBoost ...............................................................................................

FAQs

Which machine learning models are used to predict future seasoned equity offerings (SEOs)?

The thesis uses logistic regression, decision trees, random forest, and XGBoost to predict future SEOs.

How accurate are the best-performing models in predicting SEOs?

XGBoost performs best with an AUC of 0.72, while random forest achieves an AUC of 0.71, both supporting the use of nonlinear models.

Do the models provide conclusive evidence about SEO impacts on stock returns?

Additional tests using two linear regressions and one difference-in-differences estimation do not yield conclusive evidence, though existing literature suggests a negative effect.

Aug 4, 2026

Upload by Olivia Brown