Refined descriptive sampling: A better approach to Monte Carlo simulation

Tari, Megdouda; Dahmani, Abdelnasser

doi:10.1016/j.simpat.2005.04.001

Cited by 23 publications

(8 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…American options are usually solved by numerical analysis methods, including binary tree method, finite difference method and Monte Carlo simulation method (Caflisch & Chaudhary, 2004). Tari & Dahmani used to take Monte Carlo simulation to reduce the sampling bias and eliminates the problem of descriptive sampling related to the sample size (Tari & Dahmani, 2006). For options that rely on the historical price of the underlying asset to be priced, Monte Carlo simulation is the most suitable (WU & Xuan, 2006).…”

Section: Improving Of the Monte Carlo Simulationmentioning

confidence: 99%

Research on the Volatility Value of A-Share State-Owned Shipping Stocks

Hu¹,

Yu²

2022

View full text Add to dashboard Cite

The A-share stock price of state-owned shipping enterprises is higher than the discount value constantly. It implies an American call option for a speculator, so the price contains the discount value and option value. In order to help the investors to find the real value and obtain higher returns under the high volatility caused by the fluctuations of the shipping cycle or the stock market valuation cycle, Least Square Monte Carlo Simulation (LSM) method is used to calculate the implied American call option with the date of COSCO shipping energy-transportation Co., Ltd. It is found that the discount value plus volatility value among 2016-2020 is approximate to the stock price. The option value increases with the extension of maturity and the rise of expected volatility. It explains the over value and high volatility of the A-share price of cyclical stocks and small-cap stocks with poor performance.

show abstract

Section: Improving Of the Monte Carlo Simulationmentioning

confidence: 99%

Research on the Volatility Value of A-Share State-Owned Shipping Stocks

Hu¹,

Yu²

2022

View full text Add to dashboard Cite

show abstract

“…However, this slow convergence rate requires a large sample size, which hinders the application of Monte Carlo simulation when J(ξ) is relatively expensive to compute. While the convergence rate of Monte Carlo methods can be improved with different sampling techniques [20][21][22][23], these sampling strategies are still not sufficient on their own to make Monte Carlo tractable for computationally demanding analyses. In addition, propagation using Monte Carlo methods also results in noisy objective and constraint functions, which presents challenges for gradient-based optimization [24].…”

Section: A Monte Carlo Simulationmentioning

confidence: 99%

Hessian-based Dimension Reduction for Optimization Under Uncertainty

Panda

Hicken

2018

2018 Multidisciplinary Analysis and Optimization Conference

View full text Add to dashboard Cite

We present an uncertainty propagation method for computing the expected value and variance of a quantity of interest (QoI), which can then be used in a robust design optimization. To avoid intractable costs due to high-dimensional integrals, we use the Hessian of the QoI to identify the dominant nonlinear directions. Specifically, the dominant Hessian eigenmodes provide the dimensions along which the QoI is integrated in stochastic space. Explicit computation of the Hessian is avoided by using Arnoldi's method to estimate the eigenmodes. The method is applied to multi-dimensional quadratic functions and its accuracy is examined for synthetic eigenmodes.

show abstract

“…Nonlinear Programing with Non-Monotone and Distributed Line Search (NLPQLP) optimization method (Schittkowski, 2006), a well suited method for highly non-linear design spaces, was used for the optimization purpose. A descriptive sampling technique (Tari and Dahmani, 2006) was used for MCS, which is more efficient than the conventional simple random sampling method (Koch et al, 2004). 50 samples were considered for the MCS during each optimization iteration.…”

Section: Optimization Proceduresmentioning

confidence: 99%

Improvement of k-epsilon turbulence model for CFD simulation of atmospheric boundary layer around a high-rise building using stochastic optimization and Monte Carlo Sampling technique

Shirzadi

Mirzaei

Naghashzadegan

2017

Journal of Wind Engineering and Industrial Aerodynamics

View full text Add to dashboard Cite

The accuracy of the computational fluid dynamics (CFD) to model the airflow around the buildings in the atmospheric boundary layer (ABL) is directly linked to the utilized turbulence model. Despite the popularity and their low computational cost, the current Reynolds Averaged Navier-Stokes (RANS) models cannot accurately resolve the wake regions behind the buildings. The default values of the RANS models' closure coefficients in CFD tools such as ANSYS CFX, ANSYS FLUENT, PHOENIX, and STAR CCM+ are mainly adapted from other fields and physical problems, which are not perfectly suitable for ABL flow modeling. This study embarks on proposing a systematic approach to find the optimum values for the closure coefficients of RANS models in order to significantly improve the accuracy of CFD simulations for urban studies. The methodology is based on stochastic optimization and Monte Carlo Sampling technique. To show the capability of the method, a test case of airflow around an isolated building placed in a non-isothermal unstable ABL was considered. The recommended values for this case study in accordance with the optimization method were thus found to be 1.45 ≤ 1 ≤ 1.5, of 2.7 ≤ 2 ≤ 3, and 0.12 ≤ ≤ 0.15. The default value of = 1 is suggested to be acceptable while the value of is obtained through a correlation. The error of the estimated reattachment length behind the building decreased form 170% for the default values to 28% for the modified values.

show abstract

Refined descriptive sampling: A better approach to Monte Carlo simulation

Cited by 23 publications

References 14 publications

Research on the Volatility Value of A-Share State-Owned Shipping Stocks

Research on the Volatility Value of A-Share State-Owned Shipping Stocks

Hessian-based Dimension Reduction for Optimization Under Uncertainty

Improvement of k-epsilon turbulence model for CFD simulation of atmospheric boundary layer around a high-rise building using stochastic optimization and Monte Carlo Sampling technique

Contact Info

Product

Resources

About