Pavel Logacev

I solve quantitative problems for which there is no off-the-shelf solution — Demand Forecasting with sparse sales data, estimating price elasticities without controlled experiments, risk models that must stand up to decision-makers, or forecasting systems where standard approaches fail.

My methods stack: statistical modeling, Bayesian methods, time series analysis, and Causal Inference. I come from software development and academic research and can deliver what the project needs — whether that's a production-ready Python pipeline or a report for management.

I spent four years building demand and price elasticity models for hundreds of thousands of SKU-location combinations at a UK pricing consultancy (Pearson Ham Group), developed two Shopify apps for price optimization and demand forecasting, taught statistics and research methods at a teaching level, and worked in academic research with EEG and eye-tracking data. PhD in Computational Cognitive Science, 15+ years of experience in applied modeling.

Typical projects: