Predicting Success in Early-Stage Start-ups using Founding and Executive Team Characteristics

Sammanfattning

Data-driven methods have been used for investment decision support for more than two decades within the finance sector, however there are great differences in the adoption of data-driven methods in different parts of the financial market. One part of the market that has yet achieved a high level of adoption is the Venture capital (VC) industry. In this study, the use of machine learning is explored as an option to filter out high quality deals by choosing those with the highest likelihood of success. More specifically, characteristics about the founding and executive venture team is explored as independent variables for modelling the likelihood of post-seed investment within five years of the founding date.