The Market Generator

Kondratyev, Alexei; Schwarz, Christian

doi:10.2139/ssrn.3384948

Cited by 31 publications

(33 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, generative ANNs have been suggested recently as a non-parametric simulation tool for stock prices; see, for example, Henry-Labordère [2019], Kondratyev and Schwarz [2019], and Wiese et al [2019b]. Such simulation engines could then be used for option pricing and hedging, a direction still to be explored systematically.…”

Section: Further Workmentioning

confidence: 99%

Neural networks for option pricing and hedging: a literature review

Ruf¹,

Wang²

2020

JCF

View full text Add to dashboard Cite

Neural networks have been used as a nonparametric method for option pricing and hedging since the early 1990s. Far over a hundred papers have been published on this topic. This note intends to provide a comprehensive review. Papers are compared in terms of input features, output variables, benchmark models, performance measures, data partition methods, and underlying assets. Furthermore, related work and regularisation techniques are discussed.

show abstract

Section: Further Workmentioning

confidence: 99%

Neural networks for option pricing and hedging: a literature review

Ruf¹,

Wang²

2020

JCF

View full text Add to dashboard Cite

show abstract

“…Here the goal is to precisely mimic the behavior and features of historical market trajectories. This line of research has been recently pursued in e.g., Kondratyev and Schwarz (2019); Wiese et al (2019); Acciaio and Xu (2020); Bühler et al 2020; Henry-Labordère (2019).…”

Section: Generative Adversarial Approaches In Financementioning

confidence: 99%

A Generative Adversarial Network Approach to Calibration of Local Stochastic Volatility Models

Cuchiero

Khosrawi

Teichmann

2020

Risks

View full text Add to dashboard Cite

We propose a fully data-driven approach to calibrate local stochastic volatility (LSV) models, circumventing in particular the ad hoc interpolation of the volatility surface. To achieve this, we parametrize the leverage function by a family of feed-forward neural networks and learn their parameters directly from the available market option prices. This should be seen in the context of neural SDEs and (causal) generative adversarial networks: we generate volatility surfaces by specific neural SDEs, whose quality is assessed by quantifying, possibly in an adversarial manner, distances to market prices. The minimization of the calibration functional relies strongly on a variance reduction technique based on hedging and deep hedging, which is interesting in its own right: it allows the calculation of model prices and model implied volatilities in an accurate way using only small sets of sample paths. For numerical illustration we implement a SABR-type LSV model and conduct a thorough statistical performance analysis on many samples of implied volatility smiles, showing the accuracy and stability of the method.

show abstract

“…Generative models form one of the most important classes of unsupervised machine learning techniques, with numerous applications in such diverse fields as finance (generation of synthetic market data [1,2]), medicine (data anonymization [3,4]), computer vision (image-to-image translation [5]), engineering (design optimization [6]), and physics (Ising models with many-spin interactions [7,8]), to name just a few.…”

Section: Introductionmentioning

confidence: 99%

“…First, QCBMs have strictly more expressive power than classical RBMs when only a polynomial number of parameters is allowed (number of qubits in QCBM or number of visible activation units in RBM) [15]. Second, generation of an independent sample from the learned distribution can be done in a single run of the quantum circuit in the case of the QCBM -this compares favorably with up to 10 3 -10 4 forward and backward passes through the network in the case of the RBM, which are sometimes needed to achieve the state of thermal equilibrium [1]. Since a single run of the relatively deep quantum circuit can be executed in ∼10 μs (ignoring the read out and other overheads), 1 this points toward material quantum speed-up.…”

Section: Introductionmentioning

confidence: 99%

“…Second, generation of an independent sample from the learned distribution can be done in a single run of the quantum circuit in the case of the QCBM -this compares favorably with up to 10 3 -10 4 forward and backward passes through the network in the case of the RBM, which are sometimes needed to achieve the state of thermal equilibrium [1]. Since a single run of the relatively deep quantum circuit can be executed in ∼10 μs (ignoring the read out and other overheads), 1 this points toward material quantum speed-up. Third, quantum generative models can be used to load data into the quantum state, thus facilitating realizations of many promising quantum algorithms [16].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Non‐Differentiable Leaning of Quantum Circuit Born Machine with Genetic Algorithm

Kondratyev¹

2021

Wilmott

View full text Add to dashboard Cite

The Quantum Circuit Born Machine (QCBM) is a generative quantum machine learning model that can be efficiently trained and run on the NISQ‐era quantum processors. It is likely that QCBM will be one of the first quantum machine learning models to find productive applications in quantitative finance as a powerful market generator. In this paper we test QCBM performance on a dataset of spot FX log‐returns (heavy‐tailed distribution) as well as specially constructed mixture of Normal distributions (which models spiky light‐tailed distribution). The QCBM has greater expressive power than comparable classical neural networks such as restricted Boltzmann machine (RBM) and, therefore, has potential to demonstrate quantum advantage by generating high‐quality samples from the learned empirical distribution of the market risk factors while using less computational resources than its classical counterpart. However, efficient training of QCBM remains a challenging problem. Traditional differentiable learning approach may not work well when the loss function is highly non‐smooth. In such cases it may be more efficient to use the non‐differentiable learning methods. This paper proposes a non‐differentiable learning approach to the training of QCBM based on genetic glgorithm (GA). The paper also presents results of the numerical experiments which compare performance of QCBM trained with GA against performance of the equivalent classical RBM and investigates the question of GA convergence as a function of QCBM architecture and the choice of algorithm's hyperparameters.

show abstract

The Market Generator

Cited by 31 publications

References 8 publications

Neural networks for option pricing and hedging: a literature review

Neural networks for option pricing and hedging: a literature review

A Generative Adversarial Network Approach to Calibration of Local Stochastic Volatility Models

Non‐Differentiable Leaning of Quantum Circuit Born Machine with Genetic Algorithm

Contact Info

Product

Resources

About