Survey of feature selection and extraction techniques for stock market prediction

In stock market forecasting,the identification of critical features that affect the performance of machine learning(ML)models is crucial to achieve accurate stock price predictions.Several review papers in the literature have focused on various ML,statistical,and deep learning-based methods used in stock market forecasting.However,no survey study has explored feature selection and extraction techniques for stock market forecasting.This survey presents a detailed analysis of 32 research works that use a combination of feature study and ML approaches in various stock market applications.We conduct a systematic search for articles in the Scopus and Web of Science databases for the years 2011–2022.We review a variety of feature selection and feature extraction approaches that have been successfully applied in the stock market analyses presented in the articles.We also describe the combination of feature analysis techniques and ML methods and evaluate their performance.Moreover,we present other survey articles,stock market input and output data,and analyses based on various factors.We find that correlation criteria,random forest,principal component analysis,and autoencoder are the most widely used feature selection and extraction techniques with the best prediction accuracy for various stock market applications.

The effect of overseas investors on local market efficiency:evidence from the Shanghai/Shenzhen–Hong Kong Stock Connect

Using a recent stock market liberalization reform policy in China—the Stock Connect—as a quasi-natural experiment,this study examines the effect of stock market liberalization on market efficiency.Employing a dataset of 17,086 Chinese listed firms covering 2009 to 2018,we find that stock market liberalization improves the market efficiency of the Chinese mainland stock market.We further explore the potential channels through which the Stock Connect can enhance the efficiency of the A-share(A-shares refer to shares issued by Chinese companies incorporated in China's Mainland,traded in the Shanghai Stock Exchange and the Shenzhen Stock Exchange.They are denominated in Chinese RMB(the local currency).A-shares were restricted to local Chinese investors before 2003,are open to foreign investors via the Qualified Foreign Institutional Investor,RMB Qualified Foreign Institutional Investor,or the Stock Connect programs.)market.The findings show that liberalizing capital markets could benefit local market efficiency by increasing stock price informational efficiency and improving corporate governance quality.The additional analysis shows that stock market liberalization has a significant and positive impact on local market efficiency,enhancing firm value and reducing stock crash risk.We conduct various robustness checks to corroborate our findings.This study provides important policy implications for emerging countries liberalizing capital markets for foreign investors.

A Critical Review of the Effects of Stock Returns and Market Timing on Capital Structure

Capital structure is regarded as the combination of debt and equity firms used to finance operations and investments.The choice of capital structure significantly impacts a company’s cost of capital,profitability,and risk profile.Among a series of factors that affect capital structure,this paper focuses on stock returns and market timing.In this review,an array of papers is analyzed to summarize what current research claims regarding the influence of stock returns and market timing on capital structure.This paper centers on the stock return and market timing theories and also discusses other theories like the trade-off theory,the pecking order theory,and the signaling theory.

China’s Monetary Policy Impacts on Money and Stock Markets

This study investigated the impact of China’s monetary policy on both the money market and stock markets,assuming that non-policy variables would not respond contemporaneously to changes in policy variables.Monetary policy adjustments are swiftly observed in money markets and gradually extend to the stock market.The study examined the effects of monetary policy shocks using three primary instruments:interest rate policy,reserve requirement ratio,and open market operations.Monthly data from 2007 to 2013 were analyzed using vector error correction(VEC)models.The findings suggest a likely presence of long-lasting and stable relationships among monetary policy,the money market,and stock markets.This research holds practical implications for Chinese policymakers,particularly in managing the challenges associated with fluctuation risks linked to high foreign exchange reserves,aiming to achieve autonomy in monetary policy and formulate effective monetary strategies to stimulate economic growth.

Exploiting Data Science for Measuring the Performance of Technology Stocks

The rise or fall of the stock markets directly affects investors’interest and loyalty.Therefore,it is necessary to measure the performance of stocks in the market in advance to prevent our assets from suffering significant losses.In our proposed study,six supervised machine learning(ML)strategies and deep learning(DL)models with long short-term memory(LSTM)of data science was deployed for thorough analysis and measurement of the performance of the technology stocks.Under discussion are Apple Inc.(AAPL),Microsoft Corporation(MSFT),Broadcom Inc.,Taiwan Semiconductor Manufacturing Company Limited(TSM),NVIDIA Corporation(NVDA),and Avigilon Corporation(AVGO).The datasets were taken from the Yahoo Finance API from 06-05-2005 to 06-05-2022(seventeen years)with 4280 samples.As already noted,multiple studies have been performed to resolve this problem using linear regression,support vectormachines,deep long short-termmemory(LSTM),and many other models.In this research,the Hidden Markov Model(HMM)outperformed other employed machine learning ensembles,tree-based models,the ARIMA(Auto Regressive IntegratedMoving Average)model,and long short-term memory with a robust mean accuracy score of 99.98.Other statistical analyses and measurements for machine learning ensemble algorithms,the Long Short-TermModel,and ARIMA were also carried out for further investigation of the performance of advanced models for forecasting time series data.Thus,the proposed research found the best model to be HMM,and LSTM was the second-best model that performed well in all aspects.A developedmodel will be highly recommended and helpful for early measurement of technology stock performance for investment or withdrawal based on the future stock rise or fall for creating smart environments.

Predicting Stock Prices Using Polynomial Classifiers: The Case of Dubai Financial Market

Predicting stock price movements is a challenging task for academicians and practitioners. In particular, forecasting price movements in emerging markets seems to be more elusive because they are usually more volatile often accompa-nied by thin trading-volumes and they are susceptible to more manipulation compared to mature markets. Technical analysis of stocks and commodities has become a science on its own;quantitative methods and techniques have been applied by many practitioners to forecast price movements. Lagging and sometimes leading technical indicators pro-vide rich quantitative tools for traders and investors in their attempt to gain advantage when making investment or trading decisions. Artificial Neural Networks (ANN) have been used widely in predicting stock prices because of their capability in capturing the non-linearity that often exists in price movements. Recently, Polynomial Classifiers (PC) have been applied to various recognition and classification application and showed favorable results in terms of recog-nition rates and computational complexity as compared to ANN. In this paper, we present two prediction models for predicting securities’ prices. The first model was developed using back propagation feed forward neural networks. The second model was developed using polynomial classifiers (PC), as a first time application for PC to be used in stock prices prediction. The inputs to both models were identical, and both models were trained and tested on the same data. The study was conducted on Dubai Financial Market as an emerging market and applied to two of the market’s leading stocks. In general, both models achieved very good results in terms of mean absolute error percentage. Both models show an average error around 1.5% predicting the next day price, an average error of 2.5% when predicting second day price, and an average error of 4% when predicted the third day price.

Dynamics of volatility spillover between stock market and foreign exchange market: evidence from Asian Countries

Background:The purpose of this study is to examine volatility spillover effects between stock market and foreign exchange market in selected Asian countries;Pakistan,India,Sri Lanka,China,Hong Kong and Japan.This study considered daily data from 4th January,1999 to 1st January,2014.Methods:This study opted EGARCH(Exponential Generalized Auto Regressive Conditional Heteroskedasticity)model for the purpose of analyzing asymmetric volatility spillover effects between stock and foreign exchange market.Results:The EGARCH analyses reveal bidirectional asymmetric volatility spillover between stock market and foreign exchange market of Pakistan,China,Hong Kong and Sri Lanka.The results reveal unidirectional transmission of volatility from stock market to foreign exchange market of India.The analysis reveals no evidence of volatility transmission between the two markets in reference to Japan.Conclusions:The result of this study provide valuable insights to economic policy makers for financial stability perspective and to investors regarding decision making in international portfolio and currency risk strategies.

Stock Market Prediction Using Heat of Related Keywords on Micro Blog

Whether the stock market investors’ emotion can influence the stock market itself is one of the hot topic in financial research. In this paper, a method based on the heat of related keywords on Micro Blog is proposed, as Micro Blog is an ideal source for capturing public opinions towards certain topic. We choose Shanghai Composite index as the research object, through correlation analysis, Granger causality analysis, and support vector machine classification, the results have shown that the keywords heat on micro blog can make a short-time prediction of stock market, and the keyword which expresses negative emotion have more powerful prediction ability.

The impact of the COVID‑19 outbreak on Chinese‑listed tourism stocks

This research explored the effects of the coronavirus disease(COVID-19)outbreak on stock price movements of China’s tourism industry by using an event study method.The results showed that the crisis negatively impacted tourism sector stocks.Further quantile regression analyses supported the non-linear relationship between the government’s responses and stock returns.The results present that the resurgence of the virus in Beijing did bring about a short-term negative impact on the tourism industry.The empirical results can be used for future researchers to conduct a comparative study of cultural differences concerning government responses to the COVID-19.

Short-term and long-term Interconnectedness of stock returns in Western Europe and the global market

Background:The present study examines the short term dynamics and long term equilibrium relationship among the stock markets of 17 countries in Western Europe as well as the world market,using time series techniques.Methods:Weekly returns of market benchmark indices of the respective countries are used from the second week of 1995 to the fourth week of December 2013.Results:The study finds that the market returns of Austria,Belgium,the Netherlands,and France are relatively less dynamically interlinked as compared with Britain,Denmark,Finland,Germany,Portugal,Spain,Sweden,Switzerland,Greece,Ireland,Luxembourg,and Norway,which are quite dynamically interlinked within the region as well as with the MSCI world index.Conclusion:There exists a strong long run equilibrium relationship between the return distributions of the stock markets within the region.

Dynamic connectedness between stock markets in the presence of the COVID‑19 pandemic:does economic policy uncertainty matter?

This study investigates the dynamic connectedness between stock indices and the effect of economic policy uncertainty(EPU)in eight countries where COVID-19 was most widespread(China,Italy,France,Germany,Spain,Russia,the US,and the UK)by implementing the time-varying VAR(TVP-VAR)model for daily data over the period spanning from 01/01/2015 to 05/18/2020.Results showed that stock markets were highly connected during the entire period,but the dynamic spillovers reached unprecedented heights during the COVID-19 pandemic in the first quarter of 2020.Moreover,we found that the European stock markets(except Italy)transmitted more spillovers to all other stock markets than they received,primarily during the COVID-19 outbreak.Further analysis using a nonlinear framework showed that the dynamic connectedness was more pronounced for negative than for positive returns.Also,findings showed that the direction of the EPU effect on net connectedness changed during the pandemic onset,indicating that information spillovers from a given market may signal either good or bad news for other markets,depending on the prevailing economic situation.These results have important implications for individual investors,portfolio managers,policymakers,investment banks,and central banks.

General election effect on the network topology of Pakistan’s stock market: network-based study of a political event

To examine the interdependency and evolution of Pakistan’s stock market,we consider the cross-correlation coefficients of daily stock returns belonging to the blue chip Karachi stock exchange(KSE-100)index.Using the minimum spanning tree network-based method,we extend the financial network literature by examining the topological properties of the network and generating six minimum spanning tree networks around three general elections in Pakistan.Our results reveal a star-like structure after the general elections of 2018 and before those in 2008,and a tree-like structure otherwise.We also highlight key nodes,the presence of different clusters,and compare the differences between the three elections.Additionally,the sectorial centrality measures reveal economic expansion in three industrial sectors—cement,oil and gas,and fertilizers.Moreover,a strong overall intermediary role of the fertilizer sector is observed.The results indicate a structural change in the stock market network due to general elections.Consequently,through this analysis,policy makers can focus on monitoring key nodes around general elections to estimate stock market stability,while local and international investors can form optimal diversification strategies.

Herding behavior in Ramadan and financial crises: the case of the Pakistani stock market

This study examines herding behavior in the Pakistani Stock Market under different market conditions,focusing on the Ramadan effect and Crisis period by using data from 2004 to 2014.Two regression models of Christie and Huang(Financ Analysts J 51:31-37,1995)and Chang et al.,(J Bank Finance 24:1651-1679,2000)are used for herding estimations.Results based on daily stock data reveal that there is an absence of herding behavior during rising(up)and falling(down)market as well as during high and low volatility in market.While herding behavior is detected during low trading volume days.Yearly analysis shows that herding existed during 2005,2006 and 2007,while it is not evident during rest of the period.However,herding behavior is not detected during Ramadan.Furthermore,during financial crisis of 2007-08,Pakistani Stock Market exhibits herding behavior due to higher uncertainty and information asymmetry.

Principle Fluctuation Modes of the Global Stock Market

The purpose of this work is to study the principle fluctuation modes of the global stock market,which is regarded as a complex system.It is proposed that the systematic risk can be reflected by the trace calculated from the cross-correlation matrix,and the integrity can be classified into clusters according to the plus-minus signs of the elements of the eigenvectors corresponding to several top largest eigenvalues whose total value accounts for more than 60 percent of the trace.The principle fluctuation modes of 30 stock markets are in the same direction in each year of 2005-2010.According to the classification criteria proposed here,the stock markets of the Americas,Europe and Asia & Oceania are automatically classified into different clusters,while Brazil,Russia and China are separated.

Pandemic or panic?A firm-level study on the psychological and industrial impacts of COVID-19 on the Chinese stock market

This study presents a thorough investigation of the relationship between the coronavirus disease 2019(COVID-19)and daily stock price changes.We use several types of COVID-19 patients as indicators for exploring whether stock prices are significantly affected by COVID-19’s impact.In addition,using the Chinese stock market as an example,we are particularly interested in the psychological and industrial impacts of COVID-19 on the financial market.This study makes two contributions to the literature.First,from a theoretical perspective,it shows a novel quantitative relationship between the psychological response to the pandemic and stock prices.In addition,it depicts the mechanism of the shock to the stock market by pointing out the specific functional expression of the impulse reaction.To our knowledge,this is the first theoretical calculation of the impulse of a shock to the financial market.Second,this study empirically estimates the marginal effect of the COVID-19 pandemic on fluctuations in stock market returns.By controlling for stock fundamentals,this study also estimates diverse industrial responses to pandemic stock volatility.We confirm that the COVID-19 pandemic has caused panic in the stock market,which not only depresses stock prices but also inflates volatility in daily returns.Regarding the impulse of the shock,we identify the cumulative level of the pandemic variables as well as their incremental differences.As shown by our empirical results,the terms for these differences will eventually dominate the marginal effect,which confirms the fading impulse of the shock.Finally,this study highlights some important policy implications of stock market volatility and returns to work in the industry.

Markov-Switching Time-Varying Copula Modeling of Dependence Structure between Oil and GCC Stock Markets

This paper proposes a Markov-switching copula model to examine the presence of regime change in the time-varying dependence structure between oil price changes and stock market returns in six GCC countries. The marginal distributions are assumed to follow a long-memory model while the copula parameters are supposed to evolve according to the Markov-switching process. Furthermore, we estimate the Value-at-Risk (VaR) based on the proposed approach. The empirical results provide evidence of three regime changes, representing precrisis, financial crisis and post-crisis, in the dependence structure between energy and GCC stock markets. In particular, in the pre- and post-crisis regimes, there is no dependence, while in the crisis regime, there is significant tail dependence. For OPEC countries, we find lower tail dependence whereas in non-OPEC countries, we see upper tail dependence. VaR experiments show that the Markov-switching time- varying copula model performs better than the time-varying copula model.

The Mediating Effect of Corporate Governance on the Relationship Between Accounting Information and Stock Market Return of Listed Entities in Ghana

The study examined the mediating effect of corporate governance on the relationship between accounting information and stock market returns of listed entities on the Ghana Stock Exchange.The population of the study was forty(40)listed entities from 2007-2019 with 520 firm-year observations.The study applied a panel regression model that takes unobserved individual heterogeneity and distributional heterogeneity into consideration.In addition,the study employed cross-section dependence test,Levin-Lin-Chu,ImPesaran,Pesaran,Kao,and Larsson cointegration test,fully modified ordinary least square(FMOLS),and dynamic ordinary least square(DOLS).The results of unit root test showed that all the variables are integrated at first difference.Moreover,the results of cointegration test revealed that accounting information variables were cointegrated in the long run.The result of FMOLS and DOLS further revealed that all the accounting information variables with the exception of OCFPS and NTA have a direct insignificant relationship with the stock market return.The study revealed that corporate governance which was proxied by board size also strengthens the relationship between TAT and stock market return and NTA and stock market return at 5%significant level under FMOLS and DOLS respectively.

Model of Risk Forewarn and Investment Decision in Stock Markets and Its Realization

Based on the discussion of characteristic and mechanism of the stock prices volatility in Chinese emerging stock markets, this research designs an index system for risk forewarn, and builds up an investment decision model based on the forewarn of the market risk signal. Then, on probing into the structure and function of the realization of the model, the paper presents the method of data interface.