Multiple Stock Research Articles

An elicitation study to understand the equity investment motivation and decisions among Indian millennials

Purpose Millennials are a vital generational cohort of the Indian population, and understanding their motivation to participate in the stock market is crucial. This study aims to understand the investment decision-making behavior among millennials in the Indian Stock Market. Design/methodology/approach Using a cross-sectional research design that entails in-depth personal interviews, this study aims to understand the equity investment behavior of millennials. Verbatim texts from interview transcripts were used to analyze the content and arrive at themes. Findings The study investigated the motivation to enter the stock market and gained insights into how individuals make equity investment decisions considering economic and behavioral dimensions. The basis for stock selection was predominantly on the self-analysis of investors. Multiple stock selection priorities are also discussed. In addition, informants ensured asset diversification and exercised various strategies to overcome emotions. Furthermore, they suffered from various behavioral biases. Practical implications Individual investors are the least informed and most impacted stakeholders in the stock markets; therefore, this study contributes fresh insights to enhance their financial security. The paper also examines some noticeable behavioral tendencies retail investors exhibit and gathers helpful strategies for mitigating behavioral biases. Originality/value The uniqueness of the research lies in its adoption of a qualitative methodology that uses the investment experience of millennial investors to reveal the components of decision-making behavior and investor psychology. The findings are thereby unique and have significant managerial implications.

Applying ensemble learning to multiple stock price predictions: A comparative study

Predicting stock prices has been a perennial challenge and topic of interest for financial analysts, investors, and researchers alike. This holds true especially for influential tech companies that significantly impact the stock market landscape like Google and Tesla. While numerous models and methods have been applied to forecast stock prices globally, there remains a gap in a comparative, in-depth evaluation using Support Vector Regression (SVR), Recurrent Neural Network (RNN), Long Short-Term Memory network (LSTM), and XGBoost models for predicting prices of prominent tech stocks. This paper aims to bridge this gap, presenting a comprehensive model comparison for stock price forecasting. A series of datasets, inclusive of Google's stock prices and Tesla stock price, form the backbone of the analysis. Among the evaluated models, our preliminary findings indicate that LSTM and XGBoost demonstrate superior predictive capabilities, capturing intricate market dynamics with high precision. Further, the paper delves into the underlying factors and patterns driving stock prices, gleaning insights from the models' predictions. Through this research, this paper offers valuable benchmarks and insights for the best model for research in the field of stock price predictions and some evaluation of different features in the field of stock price predictions.

Two-dimensional cutting stock problem with multiple stock sizes

Two-dimensional cutting stock problem with multiple stock sizes

Bayesian semi-shared temporal modeling: A comprehensive approach to forecasting multiple stock prices

Stock prices of different companies frequently display similar temporal fluctuations because of common influencing factors. Accurate prediction of stock prices is of utmost importance for investors in determining their investment strategies. Utilizing multivariate forecasting, which involves analyzing multiple time series, has been shown to be highly effective and efficient when applied to stocks that exhibit similar temporal patterns. It is possible to model the relationship between shares by using a shared temporal model approach. Nevertheless, it is important to note that not all stocks selected for prediction demonstrate a strong correlation; certain stocks may deviate from expected patterns. Therefore, the direct implementation of a comprehensive shared temporal component model is not universally applicable. This study presents a new method called the Semi-Shared Temporal Model, which focuses on the correlation structure among variables that have similar patterns, while also modeling all stocks simultaneously. This methodology is applied to the three leading stocks of 2023: Amazon (AMZN), Alphabet (GOOG), and MercadoLibre (MELI). Based on monthly data collected from January 2010 to December 2023, the study forecasts the stock prices for the months of January to December 2024. The analysis findings suggest that the temporal patterns of AMZN and GOOG shares are highly similar, which supports the idea of modeling them together with shared temporality. Three forecasting methods are utilized: univariate models, full shared temporal models, and semi-shared temporal models. The analysis determines that the semi-shared temporal model approach produces the most precise forecasting outcomes, with a Mean Absolute Percentage Error (MAPE) of 17.97%, surpassing both univariate and full shared temporal models. The forecast for 2024 indicates a favorable trajectory for all three stocks.

Multiple Stock Prediction Based on Linear and Non-linear Machine Learning Regression Methods

In contemporary times, the pressing issue of global environmental pollution has prompted the exploration of alternative energy sources by various industries, aiming to mitigate the adverse environmental impacts caused by traditional energy production. Correspondingly, investors in the financial market have increasingly redirected their capital towards the new energy sector. Within this context, the present research endeavors to employ machine learning techniques for the prediction of Tesla's stock price. This study leverages multiple linear regression, polynomial regression, and lag models to construct models based on the datasets of TSLA, MPC, and UNG stock prices spanning the period of 2019-2020. By discerning potential patterns among these variables, the objective is to anticipate the future trajectory of TSLA stock price. According to machine learning methods, Tesla's stock price can be predicted, and the daily price of Tesla is influenced by the opening price, high price, low price and trading volume of the stock on that day. In addition, the share prices of energy companies related to Tesla also have an impact on Tesla's share price on that day. Specifically, Tesla's stock price is influenced by Natural Gas Company (UNG), which has an opposite relationship. Although common sense economics says that the crude oil market will be closely related to the new energy market. However, the results of this study demonstrated that Tesla's stock price is less influenced by Crude Oil Company (MPC).

Assessment of the impact of the economic factors on stock prices: cross-sectoral approach

This paper analyses the impact of economic factors on stock prices based on a cross-sectoral approach. After the analysis of academic literature in the field, the graphical analysis of stock price movements in different sectors is conducted and multiple stock price factors panel regression models are constructed. The results of the research show that the strongest difference between sectors was observed in the effect on stock prices of such factors as inflation and interest rates. When analyzing other factors, no significant difference in effect between sectors was observed.

New moon day anomalies of Amavasya and Muhurat trading: Gestalting the role of culture and institutions

We study the socio-cultural ecosystems, along with cross-country investments and global flow of foreign capital in a local milieu, India, having multiple stock exchanges and listed firms. India's varied investor profiles exhibit a marked preference for certain “special” days given their religious beliefs. This study tests the abstinence hypothesis to examine the “new moon day” effect, considered inauspicious by the local dominant investing community, and the duality of minds, based on India's annual festival, Diwali, considered auspicious to make investments and commence any wealth-generating activity (Muhurat trading). We find that the Indian markets generate a significant negative return on a Friday new moon day and a significant positive return on the Muhurat trading day. We propose a profit-making trading strategy that exploits the impact of the new moon day.

Implementation of Magic Formula and Acquirer’s Multiple Stock Investment Strategy in The Indonesia Stock Exchange

The Indonesia Capital Market has experienced a significant increase in investors from 2019 until now. The increase in the number of new Indonesian capital market investors is dominated by people under 30 years who prefer to invest in stocks and mutual funds. This increase in new investors does not follow by high financial literacy. Research regarding financial literacy and investment return in Indonesia showed that financial literacy affects investment return. Therefore, Indonesia’s new investors will most likely experience losses due to their lack of financial literacy. Even though there are equity funds for novice stock investors that help them minimize the error they would make if they invest themselves, they cannot choose the equity funds randomly since most equity funds cannot beat the market benchmark. This study proposed using the Magic Formula and Acquirer’s Multiple as an investment strategy for new investors in Indonesia. Both Magic Formula and Acquirer’s Multiple generated an average annual return greater than the market and Indonesia Equity Funds listed since 2016 with an average return of 26,24% and 26,32% annually. The Sharpe ratio of both methods also generated a higher ratio than the market, where Magic Formula generated an average Sharpe ratio of 0,930 annually, while Acquirer’s Multiple generated an average ratio of 1,038. The acquirer’s Multiple is recommended for novice investors because it outperforms the Magic Formula and the market in terms of actual annual returns and risk-adjusted returns.

Implementation of Magic Formula and Acquirer's Multiple Stock Investment Strategy in The Indonesia Stock Exchange

Implementation of Magic Formula and Acquirer's Multiple Stock Investment Strategy in The Indonesia Stock Exchange

Pearson correlation and transfer entropy in the Chinese stock market with time delay

Correlations between two time series, including the linear Pearson correlation and the nonlinear transfer entropy, have attracted significant attention. In this work, we studied the correlations between multiple stock data with the introduction of a time delay and a rolling window. In most cases, the Pearson correlation and transfer entropy share the same tendency, where a higher correlation provides more information for predicting future trends from one stock to another, but a lower correlation provides less. Considering the computational complexity of the transfer entropy and the simplicity of the Pearson correlation, using the linear correlation with time delays and a rolling window is a robust and simple method to quantify the mutual information between stocks. Predictions made by the long short-term memory method with mutual information outperform those made only with self-information when there are high correlations between two stocks.

Study on Comprehensive Optimization of Electric Multiple Unit Trains Passing Station Route and Rolling Stock Scheduling

In order to overcome the contradiction between the rolling stock scheduling problem of EMU at a macro level and the problem of EMU train passing station route at a micro level in some special cases, this paper comprehensively optimizes the electric multiple unit train passing station route and rolling stock scheduling. By taking the operation of the Chinese high-speed railway EMU trains as the actual background and using the actual situation that the first-level inspection and repair cycle time is twice the cycle of a high-speed railway train timetable, based on the establishment of a 48-hour continuous network, a 0-1 type integer linear programming model is constructed. While obtaining the rolling stock scheduling of EMU trains, the EMU train passing station route scheme is obtained. The experimental results show that, compared with solving the problem in two stages, the comprehensive optimization according to the method in this paper can better ensure the feasibility of the passing station route scheme and provide decision support for the actual operation.

Cutting uncertain stock and vehicle routing in a sustainability forestry harvesting problem

Sustainable forest management is concerned with the management of forests according to the principles of sustainable development. As a contribution to the field, this paper combines the Vehicle Routing Problem (VRP) (in which the vehicles are harvesters) with the Multiple Stock Size Cutting Stock Problem under uncertainty (in which the stock is logs). We present an Integer Linear Program that dynamically combines the cutting of the uncertain stock with vehicle routing, and uses it to address real-life problems. In experiments on real data from the forestry harvesting industry, we show that it outperforms a commonly used metaheuristic algorithm.

International Review of Equity Market Structure Regulation: Phase III

International Review of Equity Market Structure Regulation: Phase III

Measuring risk spillovers from multiple developed stock markets to China: A vine-copula-GARCH-MIDAS model

In order to examine the risk spillovers among multiple stock markets efficiently, we develop a vine-copula-GARCH-MIDAS model to estimate the multivariate joint distribution, and then derive CoVaR-type risk measures. Our empirical results on international stock markets show that the vine-copula-GARCH-MIDAS model is promising and is superior to several popular models. One the one hand, it exploits macroeconomic fundamentals to improve the accuracy of CoVaR measure. On the other hand, it is able to measure risk spillovers in a “multiple-to-one” pattern and solve the problem of underestimation in conventional “one-to-one” methods. What’s more, we find that there are significant risk spillovers from multiple developed stock markets including the US, Japan, and Britain, to China, which is necessary for regulators to be concerned with multiple markets simultaneously instead of a single market.

FinRL: A Deep Reinforcement Learning Library for Automated Stock Trading in Quantitative Finance

As deep reinforcement learning (DRL) has been recognized as an effective approach in quantitative finance, getting hands-on experiences is attractive to beginners. However, to train a practical DRL trading agent that decides where to trade, at what price, and what quantity involves error-prone and arduous development and debugging. In this paper, we introduce a DRL library FinRL that facilitates beginners to expose themselves to quantitative finance and to develop their own stock trading strategies. Along with easily-reproducible tutorials, FinRL library allows users to streamline their own developments and to compare with existing schemes easily. Within FinRL, virtual environments are configured with stock market datasets, trading agents are trained with neural networks, and extensive backtesting is analyzed via trading performance. Moreover, it incorporates important trading constraints such as transaction cost, market liquidity and the investor's degree of risk-aversion. FinRL is featured with completeness, hands-on tutorial and reproducibility that favors beginners: (i) at multiple levels of time granularity, FinRL simulates trading environments across various stock markets, including NASDAQ-100, DJIA, S&P 500, HSI, SSE 50, and CSI 300; (ii) organized in a layered architecture with modular structure, FinRL provides fine-tuned state-of-the-art DRL algorithms (DQN, DDPG, PPO, SAC, A2C, TD3, etc.), commonly-used reward functions and standard evaluation baselines to alleviate the debugging workloads and promote the reproducibility, and (iii) being highly extendable, FinRL reserves a complete set of user-import interfaces. Furthermore, we incorporated three application demonstrations, namely single stock trading, multiple stock trading, and portfolio allocation. The FinRL library will be available on Github at link https://github.com/AI4Finance-LLC/FinRL-Library.

Modeling and solving a real-world cutting stock problem in the marble industry via mathematical programming and stochastic diffusion search approaches

In this study, one-dimensional marble plane cutting problem is studied based on the cutting equipment productivity and effective use of marble blocks. Different types of marble planes should be cut from multiple stock sized marble blocks in parallel by using a stone block-cutting machine (gang saw). Marble blocks, which have different dimensions, and quality gradations are supplied by the marble processing factory’s own marble quarries. Different quality gradations of the marble blocks and cutting equipment productivity need to be taken into account in addition to minimize the total trim loss or cost in modelling the present cutting problem. The nature of the production process in the marble industry consists of large amount of waste. Minimizing the amount of waste can be controlled indirectly by establishing an appropriate cutting plan. Therefore, minimization of the total cost of transportation and overgrading (supplying higher quality material than specified by the customer) while determining a suitable cutting plan for the marble blocks is aimed. A mixed integer linear programming (MILP) model is developed for solving the present marble plane cutting problem for small-size instances. The salient feature of the developed model is not requiring a priori enumeration of all possible cutting patterns and to permit generation of retail to reuse. For larger size problem instances a Stochastic Diffusion Search (SDS) algorithm is developed. Extensive computational studies are performed by using test instances, which are collected from a marble processing company operating at İzmir. The computational results show that the developed models and SDS algorithm are very useful in providing effective cutting plans.

A nonparametric empirical Bayes approach to large-scale multivariate regression

Multivariate regression has many applications, ranging from time series prediction to genomics. Borrowing information across the outcomes can improve prediction error, even when outcomes are statistically independent. Many methods exist to implement this strategy, for example the multiresponse lasso, but choosing the optimal method for a given dataset is difficult. These issues are addressed by establishing a connection between multivariate linear regression and compound decision problems. A nonparametric empirical Bayes procedure that can learn the optimal regression method from the data itself is proposed. Furthermore, the proposed procedure is free of tuning parameters and performs well in simulations and in a multiple stock price prediction problem.

Chinese Super League stock prices and team performance

PurposeUtilizing the capital asset pricing model (CAPM), the purpose is to analyze whether the stock prices of the corporation that owns sport teams fluctuate based on team performance in the Chinese Super League (CSL).Design/methodology/approachSeveral CSL teams are publicly owned corporations. As such, the authors look to see if on-field performance impacts the stock price of the firms. Using the news model from previous research, seemingly unrelated regressions are estimated on CSL games from 2014 through 2017.FindingsThe results from the main models indicate some evidence of a statistical relationship between on-field team performance and stock price. Furthermore, the findings for individual teams across markets did not hold consistent across different markets. More specifically, the authors found some instances where successful on-field performance led to a decline in stock prices.Originality/valueThe present study further contributes to the growing literature related to on-field performance and stock prices. Unlike previous research, the use of the CSL as the empirical setting provides the opportunity to use multiple stock markets which provides an opportunity to further examine this relationship. Finally, the study contributes broadly to the literature on professional sports ownership structures around the world.

Set Covering Model in Solving Multiple Cutting Stock Problem

Cutting Stock Problem (CSP) is the determination of how to cut stocks into items with certain cutting rules. A diverse set of stocks is called multiple stock CSP. This study used Pattern Generation (PG) algorithm to determine cutting pattern, then formulated it into a Gilmore and Gomory model and solved by using Column Generation Technique (CGT). Set Covering model was generated from Gilmore and Gomory model. Based on the results, selected cutting patterns in the first stage can be used in the second stage. The combination of patterns generated from Gilmore and Gomory model showed that the use of stocks was more effective than Set Covering model.

Investigating Algorithmic Stock Market Trading using Ensemble Machine Learning Methods

Recent advances in the machine learning field have given rise to efficient ensemble methods that accurately forecast time-series. In this paper, we use the Quantopian algorithmic stock market trading simulator to assess ensemble methods performance in daily prediction and trading. The ensemble methods used are Extremely Randomized Trees, Random Forest, and Gradient Boosting. All methods are trained using multiple technical indicators and automatic stock selection is used. Simulation results show significant returns relative to the benchmark and large values of alpha are produced from all methods. These results strengthen the role of ensemble method based machine learning in automated stock market trading.