Startup Investment Decision Support: Application of Venture Capital Scorecards Using Machine Learning Approaches

Abstract

This research aims to explore which kinds of metrics are more valuable in making investment decisions for a venture capital firm using machine learning methods. We measure the fit of developed companies to a venture capital firm’s investment thesis with a balanced scorecard based on quantitative and qualitative characteristics of the companies. Collaborating with the management team of Rose Street Capital (RSC), we explore the most influential factors of their balanced scorecard using their retrospective investment decisions of successful and failed startup companies. Our study employs six standard machine learning models and their counterparts with an additional feature selection technique. Our findings suggest that “planning strategy” and “team management” are the two most determinant factors in the firm’s investment decisions, implying that qualitative factors could be more important to startup evaluation. Furthermore, we analyzed which machine learning models were most accurate in predicting the firm’s investment decisions. Our experimental results demonstrate that the best machine learning models achieve an overall accuracy of 78% in making the correct investment decisions, with an average of 87% and 69% in predicting the decision of companies the firm would and would not have invested in, respectively. Our study provides convincing evidence that qualitative criteria could be more influential in investment decisions and machine learning models can be adapted to help provide which values may be more important to consider for a venture capital firm.

Implications of Deep Learning for Stock Market Forecasting

“…Additionally, deep learning models have been used in the context of investment decision-making. For example, research by Bai and Zhao Bai & Zhao (2021) applies a venture capital scorecard using a machine learning approach to support startup investment decisions [48]. Additionally, research by Martinez et al (2020) showed that deep reinforcement learning can be used for early adaptive classification of temporal sequences, where agents learn to make adaptive decisions between classifying incomplete sequences now or delaying their predictions to collect more data [49].…”

Artificial Intelligence in Accounting and Finance: Challenges and Opportunities

“…Similarly, venture capital firms require new technologies to extract factors that influence the quality of their investment decisions. They employ machine learning algorithms to determine that the most important factors to consider when investing in a startup are strategic planning and team management [74].…”

Convex Least Angle Regression Based LASSO Feature Selection and Swish Activation Function Model for Startup Survival Rate

“…In this SAFLSTM, the SAF is used instead of the sigmoid activation function for enhancing the classification. Swish is motivated by the utilization of the sigmoid function for gate control in LSTM and highway networks where this SAF is expressed in the equation (11) Swish (𝑥) = 𝑥. 𝑆igmoid (𝑥). Further, the derivative expression of the swish is expressed by (12) Swish ′ (𝑥) = 𝑒 𝑥 (1+𝑒 𝑥 +𝑥) (1+𝑒 𝑥 ) 2 .…”

“…Hence, the success and failure evaluation of startups become crucial for both forthcoming and well-established entrepreneurs, and stakeholders [10]. The startup evaluation is highly subjective and preliminary-stage startups are correspondingly unpredictable as there is less amount of historical data [11]. The survival prediction of the new venture is difficult because the output is mainly based on the environmental improvements and certain complexity of each venture [12].…”