A Data-Driven Approach to Multivariate Monte Carlo Simulation

digital

Article

Magazine

STATISTICA & APPLICAZIONI

Section

Open Access

Title

A Data-Driven Approach to Multivariate Monte Carlo Simulation

Authors

Mahfuza Khatun, Sikandar Siddiqui

Publisher

Vita e Pensiero

Format

Article |

Pdf

Pages

Online from

06-2023

Doi

10.26350/999999_000050

Issn

1824-6672 (print) | 2283-6659 (digital)

Write a comment for this product

Download

Ebook format Pdf readable on these devices:

Computer
with Software:
- Adobe PDF Reader
Android
iPhone/Ipad
with App:
- Adobe PDF Reader
E-Book Reader
Ibs Tolino
Kobo

Description Readers' comments

We describe a model-free, fully data-driven approach to simulating random draws from a continuous multivariate distribution. The proposed technique is an extension of the smoothed bootstrap which explicitly accounts for local differences in the dispersion of individual data points in the sample. Results from a number of simulation experiments suggest that in many cases, the procedure presented strikes a favourable balance between the conflicting objectives of adequately reflecting key characteristics of the underlying distributions and smoothing out the gaps between the individual data points in the sample. An exemplary application indicates that the proposed procedure can be used for model selection purposes by comparing competing specifications of a given regression model with respect to their out-of-sample prediction quality.

keywords

Smooth Bootstrap, Simulation, Model Selection, Model Uncertainty

Authors biography

Mahfuza Khatun: Jahangirnagar University - Department of Finance & Banking - Savar - Dhaka - Bangladesh (e-mail: mahfuza02@juniv.edu)
Sikandar Siddiqui: Deloitte Audit Analytics GmbH - Frankfurt, German (e-mail: siddiqui@web.de)

Enter a comment for A Data-Driven Approach to Multivariate Monte Carlo Simulation