EVENT-DRIVEN VAR ANALYTICS FOR OPEN FOREIGN EXCHANGE POSITION MANAGEMENT: EMPIRICAL EVIDENCE AND A DIGITAL PLATFORM DESIGN FOR UZBEKISTAN’S BANKS

Abstract

This study evaluates whether regulatory control of open currency positions (OCPs) is sufficient to measure the economic foreign exchange risk of a commercial bank and translates the findings into requirements for a digital riskmanagement platform. Official monthly exchange rates of the Central Bank of the Republic of Uzbekistan for January 2020 – September 2025 (69 observations) are used to construct four model portfolios with the same aggregate long OCP equal to 10% of banking-system regulatory capital. Parametric, historical-simulation, and Student-t Monte Carlo VaR estimates are compared; Expected Shortfall, Kupiec backtesting, and five stress scenarios are also applied. Although every portfolio has the same regulatory position ratio, monthly 99% parametric VaR ranges from 0.22% to 1.58% of regulatory capital — a 7.24-fold difference. A 10-day square-root-of-time conversion preserves the disparity, producing estimates of approximately 0.15% – 1.09% of capital. The diversified portfolio’s VaR is 44.1% below the sum of standalone currency VaRs, demonstrating information about correlations that the additive OCP measure cannot capture. The article’s second contribution is an event-driven platform design in which every accepted exchange-rate update or transaction atomically updates the position ledger, revalues exposures, recalculates OCP, VaR, Expected Shortfall, and stress scenarios, tests both regulatory and internal limits, and stores an auditable snapshot. Manual spreadsheet-based analysis cannot provide the same calculation frequency, data consistency, or traceability. VaR should therefore complement — not replace — OCP limits, while a digital platform converts the 10-day, 99% VaR requirement of paragraph 91¹ of Regulation No. 3427 into a continuous management control.