Technical paper/Risk management
Infrequent MtM reduces neither value-at-risk nor backtesting exceptions
Frequency of repricing impacts volatility and correlation measures
Momentum transformer: an interpretable deep learning trading model
An attention-based deep learning model for trading is presented
Pricing in the gap risk of mini-futures
Mini-futures need to be priced and hedged taking sudden jumps into account
Looking beyond SA-CCR
An alternative calculation of exposure at default that handles complex portfolios is presented
Automatic implicit function theorem
New technique can improve use of adjoint algorithmic differentiation in calibration problems
Mind the gap
A default intensity model reveals the risk carried by a highly leveraged counterparty
Sec-lending haircuts and indemnification pricing
A pricing method for borrowed securities that includes haircut and indemnification is introduced
Efficient simulation of affine forward variance models
Andersen's quadratic-exponential scheme is used for simulations of rough volatility models
Nonlinear risk decomposition for any type of fund
A risk decomposition by fund manager, factor or instrument is proposed
Multi-horizon forecasting for limit order books
A multi-step path is forecast using deep learning and parallel computing
Approximating lifetime expected credit loss
Credit rating and collateral value's changes have a measurable impact on creditworthiness
NLP and transformer models for credit risk
News feeds are factored into models to predict credit events
A Darwinian theory of model risk
An ex ante methodology is proposed to analyse the model risk pattern for a broad class of structures
The Fundamental Review of the Trading Book and fat tails
Conservative capital buffers may not be enough to protect against tail events
Credit migration: generating generators
A stochastic time change helps the modelling of rating transition
Differential machine learning: the shape of things to come
A derivative pricing approximation method using neural networks and AAD speeds up calculations
Equally diversified or equally weighted?
New diversification measure enables construction of equally diversified portfolios
A new arbitrage-free parametric volatility surface
A new arbitrage-free volatility surface with closed-form valuation and local volatility is introduced
Quantifying model performance
Quality of replicating portfolio is used to measure performance of a model
Quantifying systemic risk using Bayesian networks
Creditworthiness of individual entities may offer an insight into systemic risk of financial markets
The market generator
A generative neural network is proposed to create synthetic datasets that mantain the statistical properties of the original dataset