Ed Egan

I have designed and estimated hundreds of economic models, including cross sectional, time series, panel, structural, clustering, event study, and Bayesian specifications, using Stata, MATLAB, R, SAS, SPSS, Python, Julia, and other languages. My work emphasizes causal inference, model selection, and big data analysis. I focus on reproducible workflows and scalable data architectures. I have taught database systems and led workshops on quantitative methods and SQL for big data analytics. I have also built high performance computing clusters for universities and companies to support empirical research.

My research mainly focuses on the efficiency and performance of venture capital (VC), private equity (PE), mergers and acquisition (M&A), and initial public offerings (IPOs). In 2005, I founded a data analytics firm specializing in Canadian venture capital, and until 2023 maintained several large research databases on VC, M&A, and IPOs. From 2015 to 2018, I ran an a research center focused on these topics at Rice. I developed the entrepreneurship and entrepreneurial finance curricula at Georgetown and Imperial College, and taught an executive course on IPO preparation at Imperial. I have negotiated term sheets, prepared offering materials, and worked with startups and entrepreneurship organizations for over twenty years.

I have developed machine learning (ML) methods for clustering, predictive modeling, MCMC simulation, and Bayesian inference, and have written ML libraries in MATLAB, Python, Spark, Perl, and other languages. However, I primarily apply machine learning to big data in distributed GPU-based computing environments, using CUDA and high-level frameworks such as TensorFlow and PyTorch. I also adapt traditional empirical and data science models for these contexts, and have led teams implementing ML technologies in production. I founded Petabyte Economics to evaluate new ML approaches that may support the first generation of broad-purpose, cloud-scalable economic models.

My secondary research focus is on how innovation characteristics and other factors influence intellectual property and commercialization outcomes. I developed and taught graduate courses in innovation economics at Imperial College and classes on innovation strategy at Georgetown. I contributed to the NBER patent data project and maintained near-population datasets on USPTO filings, federal grants, and other public R&D funding. I provided annual R&D portfolio analyses for five years to a Fortune 500 firm in the oil and gas sector, including assessments of patenting activity, R&D management, and joint venture performance. I also worked with two other Fortune 500 companies on innovation strategy in long-term engagements.

I worked on smartphone patent litigation for two Fortune 500 firms through Berkeley Research Group, including a multiyear engagement on patent policy advocacy, and provided data analysis to Delta Economics for a Canadian antitrust case. Much of my research evaluates policy toward entrepreneurship and innovation. I have taught courses on government and business at UBC and classes on competition, innovation policy, tax incentives, and securities regulation at UC Berkeley, Imperial, and Georgetown. As a fellow at the Baker Institute and NBER, I engaged with more than fifty public bodies and policymakers, including the Council of Economic Advisors, National Science Foundation, National Security Administration, Small Business Administration, US Cyber Command, congressional committees, state and local governments, and economic development organizations.