Entrepreneurial Decision Intelligence and Startup Survival in Volatile Markets

Abstract

This research proposes a novel hybrid of two theoretical concepts: adaptive reinforcement learning and entropy based market volatility sensing and Bayesian survival estimation, under an umbrella of entrepreneurial decision intelligence for Startup Survival (EDI-SS). When compared to the single sector/mean machine learning baselines, EDI-SS performs better than a standard mean startup survival rate of 31.4% (mean over 6 volatile sectors) and 18.7% (mean across the 6 sectors with an average of 4,095 startups per sector from 2015 to 2023). The frame work runs at a decision latency of 2.1 seconds and a score of 89% on the outcome quality, both significantly better than all other framework that has been compared. The theoretical contributions of this research will see the creation of a theoretical evolution of the concepts of decision velocity, strategic agility and longitudinal survival probability, meanwhile suggesting guidelines for the deployment of startup ecosystems, investors, and policy makers in a high-uncertainty environment.