Abstract:  Machine learning has the potential to reform public policy analysis. In a society that is becoming increasingly technology-based, this shift in methodology becomes more relevant for policymakers to make informed decisions surrounding economic interests. This study utilizes neural network vector autoregressions and long short-term memory in an attempt to predict the economic outcomes of policies surrounding public-private partnerships in Allegheny County. Public-private partnerships were a reliable source of economic revitalization after the fall of Pittsburgh’s industrial sector and the subsequent rise in their health and technology sectors, and have since become a key part of the Pittsburgh area's economy. This study evaluates the effectiveness of these partnerships, while acting as a proof of method, using neural networks and vector autoregressions to predict county-wide outcomes of policy related to these partnerships.