Stock Index Returns Research Articles

Long short-term memory autoencoder based network of financial indices

We present a novel approach for analyzing financial time series data using a Long Short-Term Memory Autoencoder (LSTMAE), a deep learning method. Our primary objective is to uncover intricate relationships among different stock indices, leading to the extraction of stock networks. We examine time series data spanning from 2000 to 2022, encompassing multiple financial crises within the S&P 500 stock indices. By training a modified LSTMAE with normalized stock index returns, we extract the inherent correlations embedded in the model weights. We create directional threshold networks by applying a fixed threshold, calculated as the sum of the mean and standard deviation of matrices from various years. Our investigation explores the topological characteristics of these threshold networks across different years. Notably, the observed network properties exhibit unique responses to the various financial crises that occurred between 2000 and 2022. Furthermore, our sector analysis reveals substantial sectoral influences during times of crisis. For example, during global financial crises, the financial sector assumes a prominent role, exerting significant influence on other sectors, particularly during the European Sovereign Debt (ESD) crisis. During the COVID-19 pandemic, the health care and consumer discretionary sectors are predominantly impacted by other sectors. Our proposed method effectively captures the underlying network structure of financial markets and is validated by a comprehensive analysis of network metrics, demonstrating its ability to identify significant financial crises over time.

A Study on the Dependency Between Selected Global Stock Markets and Gold and Silver Futures

Globalization has enabled greater volume and frequency of economic and commercial practices to be implemented worldwide. As a result, with the expansion of the scope of capital expansion, the interdependence of financial markets increases. However, negative developments such as economic crises, epidemics, wars, and uncontrolled migrations cause the relations between markets to weaken and a significant number of investors to turn to commodity investments. There are many academic studies that find that investors turn to commodity investments such as gold and silver in periods of increased economic and financial uncertainty. The aim of the study is to examine the possible correlation between selected global stock markets and gold and silver futures with time-series analyses. In the period between January 2014 and May 2024, possible relationships between the monthly returns of selected global stock indices and the monthly index returns of gold and silver futures were examined according to the Autoregressive Distributed Lag Bound Test (ARDL) Method. By calculating the error correction coefficients, it has been determined how long it takes to restore the balance in case the balance between the stock markets and commodity markets is disrupted. As a result of the research, it was determined that there is a long-term cointegration relationship between gold and silver futures index returns and selected global stock index returns. In the case of a short-term imbalance in the relationship between the yields of selected global indices and the index returns of gold and silver-term transactions, the balance was recovered within 0.9042 months for the index return of gold-term operations and 0.6549 months for the index return of silver-term operations.

A modified grey wolf optimization-based dendritic neural model for stock index return prediction

The prediction of stock index returns has always been a hot topic, which is an important basis for asset allocation and investment decisions of business managers and most investors. In order to improve the accuracy of the prediction, this paper proposes the use of a model based on the combination of the Improved Grey Wolf Optimization (IGWO) and Dendritic Neural Model (DNM). DNM has a more transparent structure, as well as a complex nonlinear processing capability and a unique architecture of automatic pruning, which can deal with the complexity of financial markets. For the traditional grey wolf optimization algorithm, this study designs four strategies to improve it. (i) Develop a nonlinear control parameter for dynamic equilibrium in global and local search algorithms. (ii) Chaos theory is introduced to plan the weights of wolves (iii) A local search scheme applicable to the head wolf is developed. (iv) The method of Levy leap is utilized to help the algorithm get rid of the local optimum. A total of 10 other algorithms are compared and validated based on data from 9 world representative stock index markets. It is pointed out that using IGWO in conjunction with DNM can achieve higher accuracy prediction.

ANALYSIS OF THE DEPENDENCIES COMMODITY PRICES AND STOCK MARKET INDEXES USING COPULA

Indonesia is rich in natural resources and occupies an important position in the global raw materials market. The country's rich resources such as oil, coal, nickel, and crude palm oil (CPO) have a significant impact on the economic situation. As one of the world's leading producers and exporters of these raw materials, Indonesia's economic fate is closely linked to price fluctuations. This study uses the copula method to model the dependence between stock and commodity returns and calculates the dependence between commodity prices (oil, coal, nickel, CPO) and Indonesian stock market index (IHSG) The data used for this analysis was sourced from Bloomberg.com, covering the period from 29 September 2021 to 29 September 2023. This study investigates the dynamic dependencies between commodity price returns and the Indonesian stock market index. The results show that the correlations between oil prices and the Indonesian stock index, and between CPO prices and the stock index are generally weak. However, there are exceptions to stock index returns, such as their relatively high dependence on coal and nickel. This diverse research provides valuable insight into the complex interdependencies in Indonesia's financial landscape. Understanding dependence between commodity prices and stock indexes is of great value to investors and policymakers, as it is the basis for making informed decisions to navigate the complex global economy.

Portfolio Selection Based on Mean-Generalized Variance Analysis: Evidence from the G20 Stock Markets

Modern finance theories have been increasingly paying attention to nonlinear and asymmetric features of stock returns. In this paper, we extend the concept of covariance to generalized covariance by using Generalized Measure of Correlation (GMC). Based on the generalized covariance which is capable of catching the nonlinearity and asymmetry in stock (index) returns, we propose a mean-generalized variance portfolio selection model which considers the gross-exposure constraint. Furthermore, we propose the corresponding nonparametric estimation approach and the global optimization algorithm to enhance the applicability of our new model. Empirical studies on G20 stock markets support that a portfolio considering nonlinear and asymmetric features among the international markets would outperform traditional ones based on mean-variance optimization and equal weighting strategy in terms of return and flexibility.

International crash risk premium

This paper investigates the international crash risk and the cross-section of stock index returns. We use the ex-ante model-free negative skewness measured by country-specific index options, proposed in Bakshi et al. (2003), as a proxy of the crash risk. We find that a country’s stock index with a high crash risk relates to a higher stock return as a risk premium across countries. The international crash risk premium exists robustly after controlling for volatility risk, macroeconomic variables, sensitivities to the international risk factors, and realized return moments. In contrast, other international risk premiums do not exist based on the exposure of such control variables. Based on the crash risk premium, we construct international stock trading strategies by sorting option-implied skewness across countries that outperform benchmark strategies by sorting the above control variables.

Social media-based implosion of Silicon Valley Bank and its domino effect on bank stock indices: evidence from advanced machine and deep learning algorithms

Social media-driven speculations play a crucial role in triggering the collapse of the banking system and stock markets. In this paper, we investigate the effect of Twitter-based investor sentiment on the collapse of Silicon Valley Bank (SVB), the 16th largest bank in the US. Additionally, we examine the spillover effect of the social media-based investor sentiment and SVB collapse on the bank stock indices from twelve countries where Global Systemically Important Banks (G-SIBs) operate. Advanced machine and deep learning models are employed to model the social media-based investors’ sentiment regarding SVB implosion and its spillover effect on the G-SIBs’ bank stock indices. Our results reveal that social media-based negative investors’ sentiment played an important role in SVB implosion. Our results further show that the negative investors’ sentiment persisted, and its systemic shock was transmitted to the G-SIBs bank stock indices. Importantly, our results provide a lead and lag relationship between investors’ sentiment and returns of G-SIBs bank stock indices. The findings of this study offer crucial insights for policymakers to consider the external shocks associated with social media-based investors’ sentiment when devising policies related to bank runs, thus helping to prevent future financial crises and cross-border contagion.

Measuring the risk and return of Indonesia's and United States Stock Index

This research investigates the relationship between the returns of selected Indonesian and US stock market indexes and their risks so as to guide new investors on how to choose their investments wisely. A quantitative descriptive method was used using performance data from three Indonesian and three US stock indexes over ten years to calculate an average return. The Sharpe Index was used to measure each index's risk. The results show that the average stock return for each index in the US is higher than the Indonesia indexes, while the level of risk in the US, on average, is lower. Investors are advised to invest in index categories with higher returns and low risk to increase the chance of gaining better returns while managing their risk to be as low as possible.

CORRELATION ANALYSIS OF THE IMPACT OF SANCTIONS ON STOCK AND INDUSTRY INDICES OF COUNTRIES INITIATING SANCTIONS AND COUNTRIES UNDER SANCTIONS

This paper presents the analysis and assessment of the impact on financial markets of the consequences of sanctions imposed at the initiative of the United States and the European Union in March 2014. One of the factors for sustainable economic development in the context of imposed sanctions is the stability of financial markets. To assess the consequences of the impact of sanctions on financial markets, we performed a correlation analysis of stock and industry indices of countries initiating sanctions and countries which are subject of sanctions, including Russia. The use of correlation analysis using current data made it possible to determine the presence, direction and strength of the relationship, and to assess the statistical significance of the obtained data in the context of leading stock and industry indices. For a detailed analysis of the impact of sanctions measures, calculations were carried out using a market model for assessing the returns of stock and industry indices, taking into account the leading factor. There is a close connection between the selected industry indices and the majority of industry indices of the countries initiating sanctions and, at the same time, there is a weak connection with the industry indices of countries subject to sanctions, including Russia. The study showed the insignificant effectiveness of the impact of sanctions measures on the Russian financial market. Based on the correlation analysis carried out, further research will allow to assess not only the impact of sanctions on the economies of the countries on which these sanctions are imposed, but also to analyze the reverse effect for the countries initiating the sanctions.

Uncertainty in Central and Eastern European markets. Evidence from Twitter-based uncertainty measures

ABSTRACT Purpose Increasing uncertainty has frequently been associated with alterations in investor behaviour in scientific discourse. Central and Eastern Europe (CEE) despite generating 10% of the EU’s GDP, has largely been overlooked in investigations into the relationship between socialmedia uncertainty and its impact on the stock markets. The purpose of this article is to examine the dependence between the largest stock indices of the CEE markets and Twitter uncertainty measures. Design/methodology/approach The Vector Autoregressive (VAR) methodology, specifically bivariate Granger causality is applied to investigate the relationship between social media messages and market behaviour reflected in returns of the main stock indices for Croatian CROBEX, Czech PX, Hungarian BUX, Polish WIG and Romanian BET indices. Findings The author discovers substantial evidence indicating that the CEE region exhibits heterogeneity concerning its relationship with uncertainty in stock markets, with measures related to recession, political events, and natural disasters being of utmost relevance. Originality/value­ While the VAR model and Granger causality have been studied in relation to many markets, there is a lack of studies on the effect of Twitter-based wider spectrum of uncertainty measures on the CEE region. Twitter-based uncertainty measures on the CEE region represent a significant knowledge gap, that when addressed, could offer valuable insight into the region’s financial dynamics, thus potentially affecting hedging strategies, enhancing investor’s awareness and informing policy-making decisions.

COVID-19 and Stock Market Behaviors: Evidence from Countries in Asia, Europe, North America and South Africa

This study explores the impact of COVID-19 on the stock market from 22nd January 2020 to 19th July 2021. We select 12 worst-hit and representative countries from 4 continents worldwide including China, Japan, Russia, India, United Kingdom, Germany, France, Italy, Spain, United States, Canada and South Africa. By using panel data model, we find both the new confirmed case and total vaccinations have negative effect on the stock index returns. The government interventions positively affect the stock market. Central banks are better to apply moderate numbers of financial tools since countries response at median level gain higher returns than countries response at low level and high level. The Coronavirus Panic Index and the Coronavirus Media Coverage Index show significant positive effect on stock returns.

YATIRIMCI DUYARLILIĞININ GÜVEN VE BEKLENTİ ENDEKSLERİ ARACILIĞIYLA İNCELENMESİ: DAVRANIŞSAL BİR YAKLAŞIM

In this study, the effect of confidence and expectation indices, which are the factors of investor sentiment, on investment decisions was investigated. In the sample, United States was chosen as an example of a developed country and Turkey was chosen as an example of a developing country. The effects of independent variables on investment instruments were examined with multiple linear regression models, and then generalized autoregressive conditional variance models were created for volatility estimates. Stock index returns (BIST100 and S&P500) and the changes in the total deposit amount in the local currencies of both countries were used as dependent variables. As independent variables, the Consumer Confidence Index, Economic Confidence Index, General Confidence Index and VIX Volatility Index were included for both countries in analyses. The monthly frequency dataset between December 2012 - April 2022 was included. As a result of the regression models, it has been observed that the independent variables generally affect the investments in both countries. As a result of the volatility models, while the independent variables did not have an effect on the investments made in the BIST100 index, it was observed that the variables related to expectations and confidence had a significant effect on the other three investment instruments.

Research on Risk Measurement in Chinese Stock Market - Based on GARCH-VaR Modeling

Chinese stock investors, whether he is an individual investor or working for a investment company, are always worried about the risk of their asset, it is difficult to measure the risk without a functional model, thus, using a model to solve it is of the pivot. The initial aim of this paper is to measure and predict the risk of Chinese stock market using GARCH-VAR model. By analyzing historical stock index return data, this paper develops a dynamic model to capture the volatility and risk correlation of the stock market. The comprehensiveness and accuracy of the results are ensured by using recent data, including stock indices that broadly cover different industries and market capitalization. The findings suggest that significant volatility and risk correlations exist in the Chinese stock market. This paper provides an effective methodology to measure and predict the level of risk in the stock market, which provides investors and risk management organizations with valuable references and decision-making bases and is important for understanding and managing the risk in the Chinese stock market, which helps stock investors to make informed optimal decisions on risk management and asset allocation.

The Effect of Seasonal Depression on Stock Market Returns

The study analyses the statistical relationship between the degree of seasonal depression and stock index returns. For this analysis, we examine the daily returns of two US and five European stock indices using OLS regression. The analysis found a statistically significant relationship between seasonal depression and the change in returns. However, due to the limited use of the reduced form, this only confirms the link between seasonal depression and returns, and further observations would be required to confirm a causal relationship.

Does Russian-Ukrainian War Impact Sustainable Investment?

PURPOSE: of the paper is to explore the reaction of sustainable investment markets to the Russian- -Ukrainian war. DESIGN/METHOD: The event study methodology was used within 30 countries, grouped by the regions – Asia, Europe, North and Central America, South America (Americas), Africa, and Oceania. The cross- -sectional regression analysis was applied to identify country-driven factors which influence SRI markets during the war. RESULTS/FINDINGS: It was confirmed that the Russian-Ukrainian war generated negative cumulative abnormal returns for 70% of the analyzed SRI markets. The cross-sectional regression analysis suggests that such country-driven factors as trade-to-GDP and pension assets-to-GDP have an impact on cumulative abnormal returns of sustainable stock indices and are influenced by the levels of masculinity/femininity dur- ing the Russian-Ukrainian war. ORIGINALITY/VALUE: These findings inform policymakers, investors, and researchers about economic and cultural factors that influence SRI markets amid negative geopolitical events. There are two main innova- tions brought by the model: a similar analysis of SRI markets during the war conducted for the first time, and the model that includes financial and cultural factors. KEYWORDS: sustainable investment, Russian-Ukrainian war, event study. JEL: G140, G150, O16.

주가와 환율 변동성에 관한 실증연구

China's tourism industry has mainly focused on inbound tourism. Starting in 1995, the Chinese government introduced a weekend holiday system, and this policy served as an opportunity to develop China's domestic tourism industry. Thanks to the golden week system introduced in 1999, domestic tourism in China has grown explosively. Accordingly, social overhead capital such as railways, aviation, and highways was rapidly expanded. From then on, the tourism industry developed simultaneously into inbound tourism, domestic tourism, and outbound tourism. It is examine the volatility spillover effect between stock returns and exchange rates for Chinese tourism companies. For the empirical analysis, we use the BEKK-GARCH model of Engle and Kroner (1995). From the empirical analysis, we find the following results. First, the shock that occurred in the previous year was found to have a statistically significant positive effect on the current stock return rate and exchange rate change rate. Second, all conditional volatility spillover effect coefficients were found to have a statistically significant positive effect. Third, a negative spillover effect was found to occur between exchange rate volatility and Shanghai Composite Stock Index returns.

The Impact of The Russia-Ukraine War on The Dynamics of The Asean-5 Stock Market

The capital market can be defined as activities related to public offerings, securities trading, public companies, institutions, and certain professions. One of the factors considered in stock investment is the Efficient Market Hypothesis. The Efficient Market Hypothesis is a financial theory that discusses how asset prices in the market already reflect all available information. This study employs an approach to detect market reaction and market efficiency based on the event of the world oil price decline due to the Russian invasion of Ukraine on March 10, 2022, through excess or abnormal return, price impact, and fluctuation patterns observed using a multifractal model. The research results on the abnormal return of the ASEAN-5 stock index showed a significant negative reaction at t-5 and t-3 during the event of the world oil price decline due to the Ukraine and Russia war in 2022 and a significant positive reaction at t-0 during the event. The average abnormal return before and after the announcement of the world oil price decline due to the Ukraine and Russia war in 2022 showed a significant negative reaction, but there was no significant difference in price impact between before and after the event. The efficiency level of the stock index in ASEAN-5 did not show a significant difference after the announcement of the world oil price decline due to the Ukraine and Russia war in 2022.

Connectedness and economic policy uncertainty spillovers to the ASEAN stock markets

Uncertainty regarding Chinese and American economic policies impacts major investable asset classes in politically contestable geographic regions such as Southeast Asia. Our objective is to examine the dependence structure of Chinese and American Economic Policy Uncertainty (EPU) and returns on six Association of Southeast Asian Nations (ASEAN) stock indices and to assess the impact of a significant statecraft consideration, the One Belt One Road (OBOR) Initiative, and an exogenous shock, the COVID-19 crisis, on the connectedness between EPU and ASEAN stock returns. We do so by employing novel connectedness and copula approaches. We find that American and Chinese EPU influence ASEAN equity indices differently across countries, with the connectedness of US EPU and ASEAN stock markets exceeding that of Chinese EPU post-2002. Further, we find that the introduction of the OBOR Initiative by China in 2013 coincided with a slight change in the dependence structure for several individual ASEAN stock indices, and show that the impact of the COVID-19 pandemic on the relationship between American and Chinese EPU and individual ASEAN stock index returns was not homogenous across countries. In doing so, we shed new light on the vulnerability of ASEAN member capital markets to statecraft considerations and exogenous shocks, as well as to economic policy decision-making in China and the United States.

Estimation of time-varying kernel densities and chronology of the impact of COVID-19 on financial markets

The time-varying kernel density estimation relies on two free parameters: the bandwidth and the discount factor. We propose to select these parameters so as to minimize a criterion consistent with the traditional requirements of the validation of a probability density forecast. These requirements are both the uniformity and the independence of the so-called probability integral transforms, which are the forecast time-varying cumulated distributions applied to the observations. We thus build a new numerical criterion incorporating both the uniformity and independence properties by the mean of an adapted Kolmogorov–Smirnov statistic. We apply this method to financial markets during the onset of the COVID-19 crisis. We determine the time-varying density of daily price returns of several stock indices and, using various divergence statistics, we are able to describe the chronology of the crisis as well as regional disparities. For instance, we observe a more limited impact of COVID-19 on financial markets in China, a strong impact in the US, and a slow recovery in Europe.

Forecasting stock index return and volatility based on GAVMD- Carbon-BiLSTM: How important is carbon emission trading?

This paper proposes a new forecast method of stock price return and volatility by GAVMD-Carbon-BiLSTM. First, based on the data from eight carbon markets in China, the logarithmic rate of return of carbon emission rights price is constructed. Second, the TVP-VAR model is used to investigate the impact effect of carbon emission rights trading, stock price return, and volatility. It is used as a predictor of return and volatility rationality. Third, the genetic algorithm is used to optimize the parameters of the variational mode decomposition. Fourth, the return and volatility of stock price are decomposed into multiple intrinsic modes, effectively reducing the data’s complexity. Finally, GAVMD is combined with BiLSTM, and the logarithmic return of carbon emission trading price is used as input to predict the stock price return and volatility. The results show that carbon emissions trading impacts the rate of return and volatility at different lead times and time points. Therefore, using it to predict stock price return and volatility can improve prediction accuracy. At the same time, combining GAVMD with BiLSTM, the prediction performance of carbon emission trading price logarithmic return rate as input is much better than other machine learning models.