id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
cord-217984-ry0z7orj	Krabichler, Thomas	Deep Replication of a Runoff Portfolio	2020-09-10		.txt	text/plain	7726	447	51	They consider a rather advanced and flexible ALM model by specifying an economic objective (e.g., the maximisation of the present value of future profits), a series of constraints (e.g., regulatory requirements and liquidity assurance) and penalty costs for constraint violations. Notice, however, that neither Markovian assumptions are needed, nor value functions or dynamic programming principles: Deep ALM will simply provide an artificial asset liability manager who precisely solves the business problem (and not more) in a convincing way, i.e. provides ALM strategies along pre-defined future scenarios and stress scenarios. An essential prerequisite for the viability of Deep ALM in a treasury department is to come up with a sufficiently rich idea of the macro-economic environment and bank-specific quantities such as market risk factors, future deposit evolutions, credit rate evolutions and migrations, stress scenarios and all the parameterisations thereof. While accounting for the regulatory liquidity constraints, a deep neural network adapts a non-trivial dynamic replication strategy for a runoff portfolio that outperforms static benchmark strategies conclusively.	./cache/cord-217984-ry0z7orj.txt	./txt/cord-217984-ry0z7orj.txt