Stephanie Droop
I am a data scientist and researcher specialising in causal inference. I am finishing a PhD at the University of Edinburgh, where my research focuses on structural causal models of free-text explanations — specifically, how to identify causal structure from natural language when some variables are unobserved. Before the PhD, I spent four years at KPMG in financial due diligence, qualifying as a chartered accountant (ACCA) and working on transactions across the UK and internationally.
My doctoral work sits at the intersection of natural language processing and causal reasoning. I build structural causal models (SCMs) under the assumption of hidden confounders — the realistic case where the variables we can observe do not tell the complete causal story. Across my research programme I have designed and run five A/B experiments and built over sixteen computational models. I write and think about how causal inference methods developed in statistics and machine learning can be applied rigorously to real decisions, not just academic benchmarks.
At KPMG I led quantitative analyses on transactions in the Deal Advisory practice. In one engagement, an order book analysis I developed identified a £50m valuation discrepancy for a client. That work gave me a ground-level understanding of the difference between correlation and causation in high-stakes settings, where decisions are costly, data is incomplete, and being wrong is expensive.
I am currently exploring roles in decision data science and applied causal inference, primarily in Edinburgh and Scotland. I am most interested in digital and technology companies where causal thinking can be embedded directly into product measurement and decision-making — not as an academic exercise, but as a practical tool for getting the right answer.