Hvad karakteriserer en vinder på aktiemarkedet? En analyse af USA’s aktiemarkedsvinderes karakteristika

Victoria Klingenberg & Emil Gudik Kristensen

Student thesis: Master thesis

Abstract

The study aims to investigate 362 American stock market winners, whose stock price have at least doubled within one calendar year from 2010 to 2018 and thereby come up with common characteristics based on, values from the annual report and stock information. Together with the 362 winners, the four major GICS industries, within the population, are analyzed separately as well. These are consumer discretionary, healthcare, industrials, and information technology, which gives a total of five populations.

The thesis consists of 6 main chapters. Chapter one to four presents the motivation, problem statement, delamination, methodology, the main theories and the data which the study and results are based on. The fifth chapter consists of two main parts 1) an analysis of the characteristics and 2) regression analysis on the characteristics and financial ratios with the aim to find a model that explain the variation in the winners stock prices. Lastly, chapter six discusses the results and ends with a conclusion.

Based on an analysis of 37 variables, the first part of the study came up with different characteristics depending on whether you look at the 362 winners as one population or on the four industries in isolation. For the total population of winners, outstanding shares, EPS, and book to market ratio were found to describe the winners. EPS and debt to equity represented the Healthcare winners, and for the Consumer Discretionary population, it was book to market ratio, ROA, and debt to equity. Further, Information Technology could be described by ROE, ROIC, and Debt to Equity. Lastly, ROA, ROE, and debt to equity were the descriptive variables of the Industrial winners.

Next, a simple regression analysis is conducted on the characteristics from the first part of the study, along with the other financial ratios against the stock return for every population. Afterwards, the significant variables were then tested in a multiple regression. A multiple regression model was found to be the best explaining model for the total population and Consumer Discretionary. However a simple model was best for the healthcare and Industrials population. Lastly, there were no significant variables for the Information technology population why no model was found.

EducationsMSc in Accounting, Strategy and Control, (Graduate Programme) Final ThesisMSc in Finance and Accounting, (Graduate Programme) Final Thesis
LanguageDanish
Publication date2020
Number of pages127