Deep Replication of a Runoff Portfolio

Abstract

To the best of our knowledge, the application of deep learning in the field of quantitative risk management is still a relatively recent phenomenon. This article presents the key notions of Deep Asset Liability Management («Deep ALM») for a technological transformation in the management of assets and liabilities along a whole term structure. The approach has a profound impact on a wide range of applications such as optimal decision making for treasurers, optimal procurement of commodities or the optimisation of hydroelectric power plants. As a by-product, intriguing aspects of goal-based investing or Asset Liability Management (ALM) in abstract terms concerning urgent challenges of our society are expected alongside. We illustrate the potential of the approach in a stylised case.

Deep learning for efficient frontier calculation in finance

“…In both cases, results reported are promising. Some cases of Asset Liability Management have been reported very recently in [KT20]. In this article, we show that neural networks are able to calculate very realistic efficient frontier in the Mean-Variance case and the Mean-CVar case.…”

Deep treasury management for banks

“…Deep ALM focuses on the problem of hedging interest rate risk of the asset and liability portfolios of banks. In the case of hedging a runoff portfolio, Krabichler and Teichmann ( 2020 ) demonstrate that their deep learning-based strategy outperforms a static replication approach as commonly used in practice. This article expands on their approach of hedging a single portfolio and applies deep stochastic control in a more comprehensive model of the ALM problem.…”

“…By parametrizing controls with neural networks, these controls can be optimized using gradient descent. This method, hereafter referred to as deep stochastic control (DSC) 5 , is the basis of deep hedging (Buehler et al, 2019 ), deep replication (Krabichler and Teichmann, 2020 ), and the Deep ALM approach developed in this article.…”

“…While most of the work on deep hedging focuses on managing financial derivatives, Krabichler and Teichmann ( 2020 ) apply the DSC approach to the problem of funding and hedging a runoff portfolio. Exemplarily, it represents a not yet unwound liability on the balance sheet of a property and casualty insurance company.…”

“…Investment decisions become high-dimensional because at each point in time, a whole series of bonds with different maturities is issued along the term structure. Krabichler and Teichmann ( 2020 ) demonstrate that applying the DSC approach in due course leads to a dynamic strategy that outperforms a static replication scheme which is commonly used in practice. Since the replication of a bond portfolio is a fundamental task within ALM on either side of the balance sheet for investment and financing decisions, the success of deep replication motivates the application of DSC to a full description of the ALM problem, which we denote as Deep ALM.…”