Analysis of stock index with a generalized BN-S model: an approach based on machine learning and fuzzy parameters

TL;DR

This paper enhances the classical BN-S model for stock index analysis by integrating fuzzy theory and machine learning, effectively capturing long-term dependence and stochastic dynamics in the S&P 500 index.

Contribution

It introduces a novel hybrid approach combining fuzzy preprocessing and machine learning to improve the BN-S model for stock index analysis.

Findings

Fuzzy parameters improve the model's ability to capture long-term dependence.

The modified model effectively represents the stochastic dynamics of the index.

The approach requires only minimal modifications to the classical BN-S model.

Abstract

In this paper we implement a combination of data-science and fuzzy theory to improve the classical Barndorff-Nielsen and Shephard model, and implement this to analyze the S&P 500 index. We pre-process the index data based on fuzzy theory. After that, S&P 500 stock index data for the past ten years are analyzed, and a deterministic parameter is extracted using various machine and deep learning methods. The results show that the new model, where fuzzy parameters are incorporated, can incorporate the long-term dependence in the classical Barndorff-Nielsen and Shephard model. The modification is based on only a few changes compared to the classical model. At the same time, the resulting analysis effectively captures the stochastic dynamics of the stock index time series.