Research preview

A recent strike on Kyiv’s Northern Pivnichnyi Bridge underscores how sudden geopolitical shocks can disrupt markets and stress‑test quantitative models. For practitioners who rely on big‑data regressions and machine learning, such events illustrate the gap between in‑sample fit and out‑of‑sample performance, and the need for robust, debiased approaches. The Shock as a Natural Experiment The destruction of two major bridges within a week creates an exogenous shock to transportation costs, supply chains, and investor sentiment. Asset prices that are sensitive to logistics – for example, commodities, regional equities, and currencies linked to Ukraine – can move sharply in the hours after the event....