Deep learning
Pricing high-dimensional Bermudan options using deep learning and higher-order weak approximation
The authors propose a deep-learning-based algorithm for high-dimensional Bermudan option pricing with the novel feature of discretizing the interval between early-exercise dates using a higher-order weak approximation of stochastic differential equations.
Neural stochastic differential equations for conditional time series generation using the Signature-Wasserstein-1 metric
Using conditional neural stochastic differential equations, the authors propose a means to improve the efficiency of generative adversarial networks and test their model against other classical approaches.
BloombergGPT: Terminal giant enters the LLM race
New large language model aims to supercharge Terminal’s ability to provide sentiment, charting and search
Momentum transformer: an interpretable deep learning trading model
An attention-based deep learning model for trading is presented
Degree of influence 2022: In the grip of volatility
Rough volatility, liquidity and trade execution were quants’ top priorities this year
Alternatives to deep neural networks in finance
Two methods to approximate complex functions in an explainable way are presented
Deep calibration of rough volatility models
Rough vol models are calibrated and fitted to SPX and Vix smiles
An end-to-end deep learning approach to credit scoring using CNN + XGBoost on transaction data
The authors find that machine learning methods can generate satisfactorily performing credit score models based on data from the 90-days prior to the score date, where traditional models can perform poorly.
Pricing barrier options with deep backward stochastic differential equation methods
This paper presents a novel and direct approach to solving boundary- and final-value problems, corresponding to barrier options, using forward pathwise deep learning and forward–backward stochastic differential equations.
Deep hedging: learning to remove the drift
Removing arbitrage opportunities from simulated data used for training makes deep hedging more robust
Quant of the year: Hans Buehler
Risk Awards 2022: Architect of deep hedging aims to supplant orthodox models with method based purely on data
Podcast: Man Group’s Zohren on forecasting prices with DeepLOB
Deep learning model can project prices around 100 ticks into the future
Degree of influence 2021: XVA marks the spot
Research into valuation adjustments is back on quants’ to-do list
Multi-horizon forecasting for limit order books
A multi-step path is forecast using deep learning and parallel computing
Deep learning profit and loss
The P&L distribution of a complex derivatives portfolio is computed via deep learning
Machines can read, but do they understand?
A novel NLP application built on a Google transformer model can help predict ratings transitions
Deep learning for discrete-time hedging in incomplete markets
This paper presents several algorithms based on machine learning to solve hedging problems in incomplete markets.
Podcast: NYU’s Kolm on transaction costs and machine learning
TCA methodologies that ignore partial fills “might be off by 20% to 30%”
NLP and transformer models for credit risk
News feeds are factored into models to predict credit events
Wells touts new explainability technique for AI credit models
Novel interpretability method could spur greater use of ReLU neural networks for credit scoring
Show your workings: lenders push to demystify AI models
Machine learning could help with loan decisions – but only if banks can explain how it works. And that’s not easy