A framework for treating model uncertainty in the asset liability management problem

The problem of asset liability management (ALM) is a classic the financial mathematics and great interest for banking institutions insurance companies. Several formulations this under various model settings have been studied Mean-Variance (MV) principle perspective. In paper, ALM revisited context uncertainty in one-stage framework. practice, issues appear to several aspects problem, e.g. process characteristics, market conditions, inflation rates, inside information effects, etc. A framework relying on notion Wasserstein barycenter presented which able treat robustly type ambiguities by appropriate handling sources (models) appropriately reformulating relevant decision making problem. proposed can be applied number different leading selection investment portfolios that remain robust uncertainties appearing market. paper concluded with numerical experiment static version employing standard modelling approaches, illustrating capabilities method very satisfactory results retrieving true optimal strategy even high noise cases.