Search results

COVID-19 evolved from a local health crisis to a pandemic and affected countries worldwide accordingly. Similarly, the impacts of the pandemic on the performance of global stock markets could be time-varying. This study applies a dynamic network analysis approaches to evaluate the evolution over time of the impact of COVID-19 on the stock markets' network.

Daily closing prices of 55 global stock markets from August 1, 2019 to September 10, 2020 were retrieved. This sample period was further divided into nine subsample periods for dynamic analysis purpose. Distance matrix based on long-range correlations was calculated, using rolling window's length of 100 trading days, rolled forward at an interval of one month's working days. These distance matrices than used to construct nine minimum spanning trees (MSTs). Network characteristics were figured out, community detection and network rewiring techniques were also used for extracting meaningful from these MSTs.

The findings are, with the evolution of COVID-19, a change in co-movements amongst stock markets' indices occurred. On the 100th day from the date of reporting of the first cluster of cases, the co-movement amongst the stock markets become 100% positively correlated. However, the international investor can still get better portfolio performance with such temporal correlation structure either avoiding risk or pursuing profits. A little change is observed in the importance of authoritative node; however, this central node changed multiple times with change of epicenters. During COVID-19 substantial clustering and less stable network structure is observed.

It is confirmed that this work is original and has been neither published elsewhere, nor it is currently under consideration for publication elsewhere.

The purpose of this study is to develop a precise Islamic securities index forecasting model using artificial neural networks (ANNs).

The data of daily closing prices of KMI-30 index span from Aug-2009 to Oct-2019. The data of 2,520 observations are divided into training and test data sets by using the 80:20 ratio, which corresponds to 2016 and 504 observations, respectively. In total, 25 features are used; however, in model selection step, based on maximum accuracy, top ten indicators are selected from several iterations of predictive models.

The results of feature selection show that top five influencing indicators on Islamic index include Bollinger Bands, Williams Accumulation Distribution, Aroon Oscillator, Directional Movement and Forecast Oscillator while Mesa Sine Wave is the least important. The findings show that the model captures much of the trend and some of the undulations of the original series.

The findings of this study may have important implications for investment and risk management by using index-based products.

Numerous studies proved that traditional econometric techniques face significant challenges in out-of-sample predictability due to model uncertainty and parameter instability. Recent studies show an upsurge of interest in machine learning algorithms to improve the prediction accuracy.