A0290
Title: Cointegration analytics for production and risk hedging
Authors: Tingjin Yan - East China Normal University (China) [presenting]
Jinhui Han - Peking University (China)
Abstract: A production planning problem is studied where the production decision must be made at the start of the planning horizon, prior to the realization of random demand, while a financial hedging strategy is implemented during the planning horizon to mitigate aggregate risk. A new approach is proposed, leveraging cointegration analytics to model the comovement between demand and financial series. Using real-world data, the cointegration relationship is validated, and a cointegration factor is constructed, expressed as a linear combination of demand and financial series. Based on the mean-reverting nature of the cointegration factor, a time-consistent hedging strategy is developed, and the optimal production capacity is characterized within a mean-variance framework that balances return and risk. The cointegration intensity, defined as the mean-reverting speed of the cointegration factor, acts as a key indicator of hedging potential. In general, a higher cointegration intensity leads to more effective hedging under moderate risk tolerance, granting the firm greater operational flexibility to expand production. The hedging strategy adjusts either in the same or opposite direction as the cointegration factor fluctuates and reverts to its mean, depending on the relative weight placed on risk minimization versus profit enhancement.
