Leveraging Econometric and Deep Learning Methods for Comprehensive Analysis of Global Equity Market Returns and Volatility

Abstract

Prokarsha Kumar Ghosh

In this paper, a comprehensive analytical study was carried out between the MSCI Global Index and the iShares MSCI ACWI ETF to assess similarities, differences, and predictive behaviour in global market performance. Historical financial data for both indices were gathered, pre-processed, and examined using a combination of statistical techniques, visual analysis, and predictive modelling methods. Descriptive statistical summary measures were analysed to identify key performance indicators such as average returns, volatility, correlation, and risk-adjusted metrics, while inferential statistical tests were employed to evaluate the significance of performance variations over time. Various visualisation models were developed to illustrate the major factors driving fluctuations in both indices, allowing key contributors within the dataset to be identified. Regression analyses, including least squares and Bayesian methods, were applied to investigate index movements in relation to selected macroeconomic indicators, with model accuracy being validated through standard evaluation metrics. In addition, deep learning approaches, particularly Long Short-Term Memory (LSTM) networks, Deep Neural Networks (DNN), and other neural network models, were implemented to capture nonlinear patterns and temporal dependencies present in financial time-series data, resulting in enhanced forecasting performance compared with traditional models. The findings indicated that the MSCI Global Index and ACWI ETF are characterised by a strong positive correlation and a high degree of alignment in representing global market trends, while exhibiting distinct performance differences attributable to variations in regional and sectoral weightings. The combined application of statistical, machine learning, and deep learning techniques enabled deeper insights into index dynamics and highlighted the effectiveness of advanced analytical methods in supporting global portfolio evaluation and strategic investment decision-making.

PDF