The Prediction of Venture Capitalists' Investment Propensity Using Machine Learning

Abstract

This paper describes the most visible data science methods suitable for entrepreneurial research and provides links to literature and big data resources for venture capitalists. In the results, first, all organizational characteristics such as the characteristic of parent company of VC, the fund size of VC, and the reputation of VC, have significant influences on the risk-taking investment of venture capitalists, while functional background, school prestige, and VC experience except educational level among individual characteristics have significant influences on the risk-taking investment of venture capitalists. Second, for the full model, the accuracy rate is 0.855, which implies that the error rate is 0.145. Among the venture capitalists who are predicted not to do risk-taking investment, the accuracy that would not do risk-taking investment is 85.75%, and the accuracy that do risk-taking investment is 79.59% among the venture capitalists who are predicted to do risk-taking investment.