We present an actuarial loss reserving technique that takes into account both claim counts and claim amounts. Separate (over-dispersed) Poisson models for the claim counts and the claim amounts are combined by a joint embedding into a neural network architecture. As starting point of the neural...

We introduce two neural network models designed for application in statistical learning. The mean-variance neural network regression model allows us to simultaneously model the mean and the variance of a response variable. In case of a two-dimensional response vector, the...

The main goal of this paper is to better understand the behavior of credit spreads in the past and the potential risk of unexpected future credit spread changes. One important consideration to note regarding credit spreads is the fact that bond spreads contain a liquidity premium, which...

The main idea of this paper is to embed a classical actuarial regression model into a neural network architecture. This nesting allows us to learn model structure beyond the classical actuarial regression model if we use as starting point of the neural network calibration exactly the classical...

This paper will outline the functionality available in the CovRegpy package for actuarial practitioners, wealth managers, fund managers, and portfolio analysts written in Python 3.7. The major contributions of CovRegpy can be found in the CovRegpy_DCC.py, CovRegpy_IFF.py, CovRegpy_RCR.py,...

Forecast combinations, also known as ensemble models, routinely require practitioners to select a model from a massive number of potential candidates. Ten explanatory variables can be grouped into 21078 forecast combinations, and the number of possibilities increases further to 21078+21078 if we...

Forecast combinations, also known as ensemble models, routinely require practitioners to select a model from a massive number of potential candidates. Ten explanatory variables can be grouped into 21078 forecast combinations, and the number of possibilities increases further to 21078+21078 if we...

We present a data-driven proof of concept model capable of reproducing expected counterparty credit exposures from market and trade data. The model has its greatest advantages in quick single-contract exposure evaluations that could be used in front office xVA solutions. The data was generated...

GLOBAL FINANCE LIQUIDITY RISK REVISITED: Development of A Framework for Liquidity Assessment in Portfolio Construction Process: Presentations to the JP Morgan Global Head of Quant Research & Analytics and US Head of Portfolio Construction Teams:Presentations To: JP Morgan Global Head of Quant...

GLOBAL FINANCE LIQUIDITY RISK REVISITED: JP Morgan Alternative Assets Portfolio Liquidity Assessment Framework & Models: $500 Billion Fund of Funds: 17 Asset ClassesPresentations atJP Morgan World HQ, 270 Park Ave, Manhattan, NY, USAToJP Morgan Global Head of Quant Research & Analytics, JP...