Nonlinear Regression Analysis by Global Optimization: A Case Study in Space Engineering

Pintér, János D.; Castellazzo, Alessandro; Vola, Mariachiara; Fasano, Giorgio

doi:10.1007/978-3-319-41508-6_11

Cited by 2 publications

(1 citation statement)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As calibration is an optimization problem, its implementation is identified by the choice of its objective function, its constraints and the algorithm used. Algorithms for such global minimization problems are heuristic methods of mainly three types, i.e., stochastic search, multi-start and surrogate (see e.g., Pintér (2006) and Kochenderfer and Wheeler (2019) textbooks), while objective function can be modified to include penalty terms if a regularization of the problem is desired (see e.g., Tikhonov (1963)). These choices are necessarily model-dependent, as each log-return distribution can lead to different levels of non-convexity of the objective function.…”

Section: Introductionmentioning

confidence: 99%

Multivariate Lévy Models: Calibration and Pricing

Amici¹,

Brandimarte²,

Messeri³

et al. 2023

Preprint

View full text Add to dashboard Cite

In this research we employ a range of multivariate asset models based on Lévy processes to price exotic derivatives. We compare their ability to fit market data and replicate price benchmarks, and evaluate their flexibility in terms of parametrization and dependence structure. We review recent risk-neutral calibration approaches and techniques in the multivariate setting, and provide tools to make well-informed decisions in a practical context. A special focus is given to the ability of the models to capture linear and nonlinear dependence, with implications on their pricing performance. Given the exotic features of the analyzed derivatives, valuation is carried out through Monte Carlo methods.

show abstract

Section: Introductionmentioning

confidence: 99%