As historical data are typically unavailable for a start-up, risk assessment is always complex and challenging. Traditional methods are incapable of capturing all facets of this complexity; therefore, more sophisticated tools are necessary. Using an expert-elicited Bayesian networks (BNs) methodology, this paper aims to provide a method for combining diverse sources of information, such as historical data, expert knowledge, and the unique characteristics of each start-up, to estimate the default rate at various stages of the life cycle. The proposed method not only reduces the cognitive error of expert opinion for a new start-up but also considers the learning feature of BNs and the effect of lifespan when updating default estimations. In addition, the model considers the impact of investors’ risk appetite. Furthermore, the model can rank the most effective risk factors at various stages. The receiver operating characteristic (ROC) curve was utilized to assess the model’s explanatory power. Moreover, three distinct case studies were used to demonstrate the model’s capabilities.
