Stock Selection Research Articles

Application of the AHP-TOPSIS Method to Support Stock Investment Decisions Based on Financial Ratio Analysis

Stock is one form of investment that is currently popular, both for young people, parents, and industry. This can be seen from the increasing number of investors and companies listed on the Indonesia Stock Exchange (IDX). This makes investors confused in determining the best stock choices. Analysis of stock selection needs to be done before someone invest so that the selected stock does not lose and can generate optimal profits. This study discusses the recommendation for alternative stocks from the IDX30 index. The parameters for selecting alternative stocks considered include the criteria of Earning Per Share (EPS), Return on Assets (ROA), Return on Equity (ROE), and Net Profit Margin (NPM). Then the weight of each criterion will be searched using the Analytical Hierarchy Process (AHP) method and sorted using the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) method to produce stock recommendations from companies listed on the Indonesia Stock Exchange. The results showed that the Earning Per Share (EPS) criteria became the first priority with a weight of 0,2927, then Return on Equity (ROE) with a weight of 0,2728, followed by Net Profit Margin (NPM) with a weight of 0,2583, and Return on Assets (ROA) with a weight 0,1759. Then alternative results are obtained based on the preference value and ranking 14 companies that have a preference value above 0.5 and can be used as a consideration in making investment decisions, with BBCA as a priority alternative.

Analysis of Multifactor Fundamentals Stock Selection Based on Backtesting

In recent years, quantitative finance has become a major trend for investing which brings stable returns with controllable risks. Among various different quantitative strategies, multifactorial stock selection strategy based on fundamental data (e.g., financial statements, macro- and micro-economy data) is one of the widely investigated strategies. On this basis, this study chooses Chinese listed company to verify the feasibility and effectiveness of the stock selection strategy. To be specific, the Ricequant platform is utilized to realize the backtesting as well as data retrieving in order to estimate and evaluate the performances of the strategies. According to the analysis, several indicators show great ability to gain extra returns compared with systematic risks and market performances. In other words, the feasibility of explicability of the quantitative strategy based on multifactorial model is verified in Chinese market. Overall, these results shed light on guiding further exploration of fundamental analysis of different underlying assets based on multifactorial analysis.

PoolParty2: An integrated pipeline for analysing pooled or indexed low-coverage whole-genome sequencing data to discover the genetic basis of diversity.

Whole-genome sequencing data allow survey of variation from across the genome, reducing the constraint of balancing genome sub-sampling with estimating recombination rates and linkage between sampled markers and target loci. As sequencing costs decrease, low-coverage whole-genome sequencing of pooled or indexed-individual samples is commonly utilized to identify loci associated with phenotypes or environmental axes in non-model organisms. There are, however, relatively few publicly available bioinformatic pipelines designed explicitly to analyse these types of data, and fewer still that process the raw sequencing data, provide useful metrics of quality control and then execute analyses. Here, we present an updated version of a bioinformatics pipeline called PoolParty2 that can effectively handle either pooled or indexed DNA samples and includes new features to improve computational efficiency. Using simulated data, we demonstrate the ability of our pipeline to recover segregating variants, estimate their allele frequencies accurately, and identify genomic regions harbouring loci under selection. Based on the simulated data set, we benchmark the efficacy of our pipeline with another bioinformatic suite, angsd, and illustrate the compatibility and complementarity of these suites using angsd to generate genotype likelihoods as input for identifying linkage outlier regions using alignment files and variants provided by PoolParty2. Finally, we apply our updated pipeline to an empirical dataset of low-coverage whole genomic data from population samples of Columbia River steelhead trout (Oncorhynchus mykiss), results from which demonstrate the genomic impacts of decades of artificial selection in a prominent hatchery stock. Thus, we not only demonstrate the utility of PoolParty2 for genomic studies that combine sequencing data from multiple individuals, but also illustrate how it compliments other bioinformatics resources such as angsd.

Quantitative Multi-factor Stock Selection Strategy Based on Internet Service Industry

As a matter of fact, with regard to stock investment practice, quantitative strategies have become popular with the rapid development of China's capital market. Plenty of quantitative strategies have been proposed and implemented into the Chinese markets, where multi-factor stock selection is one of them. Based on the multi-factor model commonly used in quantitative investment, this article selects relevant data of the Internet service industry in the A-share market from 2018 to 2022, conducts an industry-level empirical test on 15 factors in 5 categories that are widely used, and constructs investments accordingly combination. According to the analysis, the research found that the stock portfolio constructed based on the above method can outperform the market, far exceeding the performance of the CSI 300 during the same period, indicating that the multi-factor model is still effective in the Internet service industry, and this industry has high investment value. Overall, these results shed light on guiding further exploration of quantitative analysis of stock market.

Large-Scale Group Decision-Making Method Using Hesitant Fuzzy Rule-Based Network for Asset Allocation

Large-scale group decision-making (LSGDM) has become common in the new era of technology development involving a large number of experts. Recently, in the use of social network analysis (SNA), the community detection method has been highlighted by researchers as a useful method in handling the complexity of LSGDM. However, it is still challenging to deal with the reliability and hesitancy of information as well as the interpretability of the method. For this reason, we introduce a new approach of a Z-hesitant fuzzy network with the community detection method being put into practice for stock selection. The proposed approach was subsequently compared to an established approach in order to evaluate its applicability and efficacy.

Combination of RNAseq and RADseq to Identify Physiological and Adaptive Responses to Acidification in the Eastern Oyster (Crassostrea virginica).

Ocean acidification (OA) is a major stressor threatening marine calcifiers, including the eastern oyster (Crassostrea virginica). In this paper, we provide insight into the molecular mechanisms associated with resilience to OA, with the dual intentions of probing both acclimation and adaptation potential in this species. C. virginica were spawned, and larvae were reared in control or acidified conditions immediately after fertilization. RNA samples were collected from larvae and juveniles, and DNA samples were collected from juveniles after undergoing OA-induced mortality and used to contrast gene expression (RNAseq) and SNP (ddRADseq) profiles from animals reared under both conditions. Results showed convergence of evidence from both approaches, particularly in genes involved in biomineralization that displayed significant changes in variant frequencies and gene expression levels among juveniles that survived acidification as compared to controls. Downregulated genes were related to immune processes, supporting previous studies demonstrating a reduction in immunity from exposure to OA. Acclimation to OA via regulation of gene expression might confer short-term resilience to immediate threats; however, the costs may not be sustainable, underscoring the importance of selection of resilient genotypes. Here, we identified SNPs associated with survival under OA conditions, suggesting that this commercially and ecologically important species might have the genetic variation needed for adaptation to future acidification. The identification of genetic features associated with OA resilience is a highly-needed step for the development of marker-assisted selection of oyster stocks for aquaculture and restoration activities.

The journey is the reward: A study of corporate site visits and mutual fund performance

Institutional investors often visit corporate sites to conduct first-hand assessments of firms' business operations and management quality. We find that corporate site visits contribute significantly to mutual fund performance in China. Our results indicate that corporate site visiting is an important part of mutual funds' stock selection process. Fund managers utilize these visiting activities as a verification channel to affirm their selection of outperforming stocks. The impact on fund performance is stronger when funds are active, have more concentrated equity holdings or invest in small-cap stocks. Performance enhancement is more profound when mutual funds visit firms with high information disclosure quality, and when fund managers have an alumni connection with corporate executives.

Portfolio management using Additive Ratio Assessment based stock selection and deep learning for prediction

Portfolio management using Additive Ratio Assessment based stock selection and deep learning for prediction

Correlation-based investment strategies: A comparison between Chinese and US stock markets

This paper studies the feasibility of correlation-based investment strategies in the Chinese and US stock markets. After testing eight different correlations between returns and various factors under two different stock selection methods, the highest correlation is identified between returns and volatility under the Pearson correlation coefficient, in which the volatility is defined as the ratio of the differences between the maximum and minimum prices within two successive trading days. The investment strategies are investigated under two stock selection methods and three portfolio construction methods, and SSCI and S&P500 Index are chosen as benchmarks respectively. In the Chinese stock market the contrarian strategy obtain significant positive excess returns while in the US stock market the momentum strategy perform better. The results show that correlation-based investment strategy is more feasible in the Chinese stock market than in the US, and the US stock market is more efficient than the Chinese stock market.

Determinant Performance of Islamic Equity Funds in Indonesia

One of the issues that have an impact on the growth of Islamic Equity Funds is financial risk tolerance. While it is a fundamental issue that is directly related to Islamic Equity Fund’s manager performance. If investment managers perform better to cultivate Islamic Equity Funds, then the intention in investing within Islamic Equity Funds will be increased. This research aims to determine and provide information about Islamic Equity Funds' performance in Indonesia. It used quantitative statistical analysis to investigate the influence of stock selection skills, market timing ability, fund size, fund age, expense ratio, and portfolio turnover towards the Islamic Equity Funds performance. This study used annual data over the periods 2015-2020 which was obtained from the prospectus of each Islamic Equity Fund, Central Bank Indonesia (BI), and Statistics Indonesia (BPS). The result indicates that stock selection skill and expense ratio have a positive significant impact on the performance of Islamic Equity Funds, while market timing ability, fund age, and fund size have not a significant impact but the result revealed positive values. Otherwise, the portfolio turnover has a negative significant impact on Islamic Equity Funds' performance. Hence, it can be inferred that the higher investment managers ability to select its portfolio stock, the better return will be obtained and it can be inferred that the expense ratio of an equity fund reflects how much it pays for portfolio management, administration, marketing, and distribution, among other things. Originality/Value: This paper focuses on the issue that occur in Islamic equity funds which are very essential as an investment portfolio, and only a few papers have conducted research in this area. Therefore, knowing the key factors that influence the performance of Islamic equity funds can optimize their function and attract investors.

Multi-factor Stock Selection Strategy Incorporating Liquidity Risk

As an emerging market dominated by retail investors, China's stock market is greatly affected by external shocks such as market noise and policy changes, and the traditional fundamental factors have limited ability to predict the future returns of stocks. Stock liquidity is also affected by noise factors and macro factors, which can be regarded as one of their external performance. Based on the multi-factor stock selection model, this paper uses the 2010-2022 CSI 300 stock data to construct a fundamental fusion factor and another fusion factor considering both fundamental and liquidity risk factors, then compares the stock selection effect of the two fusion factors through a variety of test indicators. The empirical results show that the IC value of the fusion factor considering both fundamental and liquidity risk factors is 0.0267, which is greatly improved compared with the fundamental fusion factor with IC value of only 0.0195, and its excess yield of 64.728% is 13.134% higher than the fundamental fusion factor, while the benchmark win rate is increased by nearly 3% and the maximum retracement is reduced by nearly 12%. The research shows that the effectiveness and stability of the fusion factor after adding the liquidity index to predict the future return of the stock and obtain the excess profits are better than the fundamental fusion factor. Therefore, considering the measurement and inclusion of stock liquidity level can improve the yield of stock selection strategy in the Chinese stock market to a certain extent.

Hedge fund manager timing and selectivity skill over time. A holdings-based estimate

We study the performance of hedge fund firms from 1999 through 2017, employing a holdings-based approach to decompose overall performance into stock selection and three distinct timing abilities: market return, volatility, and liquidity. In particular, we introduce a liquidity timing ability for the first time using a holdings-based measure. In aggregate, we find evidence for stock picking skill that diminishes over time, but little evidence of timing ability. We find weak evidence that managers exhibit persistence in selectivity skill. At the individual hedge fund firm level, bootstrap analysis suggests that the top managers’ selectivity skill can be separated from luck.

A generalization of multi-source fusion-based framework to stock selection

Selecting outstanding technology stocks for investment is challenging. Specifically, the research of the investment for academic purposes is not mature enough due to the disarray of the publications and the overwhelming informative experience from profit-making websites. Often, some authors entitle the stock price prediction as a stock selection problem; both prediction and selection are just sub-sections of portfolio management. Moreover, stock websites provide numerous potential criteria showing various evaluations to simulate and monitor the stock market professionally, which increases the difficulty of academic studies on stock selection.The paper generalizes a novel framework with a user-interactive interface for stock selection problems based on multi-source data fusion and decision-level fusion to enhance reliability limited by the narrow criteria performance and the strength of a model overcoming the weakness of a single-performed model. This framework benefits the time-series prediction and decision-making study. Besides, we propose adopting dynamic time warping to assist a task-learning process by customizing a loss function that improves the accuracy of data prediction. The experiment shows that the proposed method reduces the prediction log error by 6.3% on average and decreases the warping cost by 5.6% on average over all cases of real-situation data. Finally, we illustrate the proposed framework by implementing it in a real-world stock data selection. The results are practical and effective, further justified through a detailed ablation study. The source code will be available at https://github.com/lingping-fuzzy.

Retracted: Fuzzy Decision-Making Analysis of Quantitative Stock Selection in VR Industry Based on Random Forest Model

Retracted: Fuzzy Decision-Making Analysis of Quantitative Stock Selection in VR Industry Based on Random Forest Model

An analysis of the applicability of Markowitz Model in China

This research project aims to analyze whether Markowitz Model fits with the Chinese market and whether it can be applied to the Chinese market. The logical chain of this paper is to verify the rationality and stability of the stock selection strategy in the Chinese market by comparing it with the market benchmark and related indexes, and finally verify whether the Markowitz Model applies to the Chinese market with the conclusion of whether the stock selection strategy is reasonable and stable. I selected 20 representative stocks among the CSI 300 stocks and used Markowitz Model to conduct linear regression. Then I used the stock selection strategy to update the stock selection to maximize benefits regularly and compared it with the CSI 300 index to verify the similarity and correlation between it and CSI 300. Finally, the conclusion is that Markowitz's Model fits the Chinese market.

Stock picking with machine learning

AbstractWe analyze machine learning algorithms for stock selection. Our study builds on weekly data for the historical constituents of the S&P500 over the period from January 1999 to March 2021 and builds on typical equity factors, additional firm fundamentals, and technical indicators. A variety of machine learning models are trained on the binary classification task to predict whether a specific stock outperforms or underperforms the cross‐sectional median return over the subsequent week. We analyze weekly trading strategies that invest in stocks with the highest predicted outperformance probability. Our empirical results show substantial and significant outperformance of machine learning‐based stock selection models compared to an equally weighted benchmark. Interestingly, we find more simplistic regularized logistic regression models to perform similarly well compared to more complex machine learning models. The results are robust when applied to the STOXX Europe 600 as alternative asset universe.

Correlations of betanodavirus load in the brain of experimentally infected sea bass (Dicentrarchus labrax, L.) with varying levels of resistance to viral nervous necrosis

The red-spotted grouper nervous necrosis virus (RGNNV) genotype of Betanodavirus causes significant losses to the European sea bass aquaculture industry. The lack of widely available vaccines against this genotype has led researchers and the industry to investigate other avenues for reducing mortality caused by the disease, such as the selection of resistant stocks. The objectives of this study were: a) to assess the presence of natural resistance in sea bass families propagated in the genetic improvement program of Nireus S.A., Greece and, b) to test an ELISA method able to detect viral coat antigen and measure viral load in the brain of tested fish. The second objective was to investigate virus antigen fluctuations during the course of the infection and its correlation with disease development, mortality and resistance. Further aspects of disease pathogenesis were investigated. A population of sea bass consisting of 89 families was experimentally infected with an RGNNV genotype of Betanodavirus; mortalities were recorded and brain samples from dead and survived fish were collected. Experimental infection of fish resulted in a typical pattern of mortality development that reached 56%. Different levels of natural resistance between families were found with cumulative mortality ranging from 20% to 86.2%. There was no statistical difference between the weight of dead fish and the resistant and susceptible families, except when extreme phenotypes were tested, indicating that the weight of fish that died was not a significant factor of final mortality. Brain virus load in the population as a whole increased sharply until D6 post-infection and then gradually dropped until the end of the experiment. When the viral load in the population (dead and survived fish) was tested against time of death, no correlation was found. In fish that survived the infection, the virus load remained high. The viral load per mg of brain tissue in samples taken from dead fish was not a factor that influenced family cumulative % mortality. Brain samples from survived fish reproduced the disease after infection of healthy fish. This study revealed a natural resistance of sea bass to Betanodavirus infection. This important finding can be used as an additional tool in reducing the mortality of cultured stocks. Additionally, there was clear evidence that fish surviving infection become subclinical carriers of the pathogen. Apparently, survival may not be associated with mechanisms of viral clearance but to other mechanisms that may suspend viral replication or aid infected cell endurance. These findings may assist in developing future vaccines and stock selection. A combination of effective vaccination and resistant stocks could be the way ahead.

Stock Selection Using Roy Criteria to Construct a Portfolio and the Effects of Variables on Portfolio Return

Stock Selection Using Roy Criteria to Construct a Portfolio and the Effects of Variables on Portfolio Return

Phenotypic and genetic characterization of Africanized Apis mellifera colonies with natural tolerance to Varroa destructor and contrasting defensive behavior.

Africanized Apis mellifera colonies with promising characteristics for beekeeping have been detected in northern Argentina (subtropical climate) and are considered of interest for breeding programs. Integral evaluation of this feral material revealed high colony strength and resistance/tolerance to brood diseases. However, these Africanized honeybees (AHB) also showed variable negative behavioral traits for beekeeping, such as defensiveness, tendency to swarm and avoidance behavior. We developed a protocol for the selection of AHB stocks based on defensive behavior and characterized contrasting colonies for this trait using NGS technologies. For this purpose, population and behavioral parameters were surveyed throughout a beekeeping season in nine daughter colonies obtained from a mother colony (A1 mitochondrial haplotype) with valuable characteristics (tolerance to the mite Varroa destructor, high colony strength and low defensiveness). A Defensive Behavior Index was developed and tested in the colonies under study. Mother and two daughter colonies displaying contrasting defensive behavior were analyzed by ddRADseq. High-quality DNA samples were obtained from 16 workers of each colony. Six pooled samples, including two replicates of each of the three colonies, were processed. A total of 12,971 SNPs were detected against the reference genome of A. mellifera, 142 of which showed significant differences between colonies. We detected SNPs in coding regions, lncRNA, miRNA, rRNA, tRNA, among others. From the original data set, we also identified 647 SNPs located in protein-coding regions, 128 of which are related to 21 genes previously associated with defensive behavior, such as dop3 and dopR2, CaMKII and ADAR, obp9 and obp10, and members of the 5-HT family. We discuss the obtained results by considering the influence of polyandry and paternal lineages on the defensive behavior in AHB and provide baseline information to use this innovative molecular approach, ddRADseq, to assist in the selection and evaluation of honey bee stocks showing low defensive behavior for commercial uses.

Volatility Prediton based on Multiple-factor Model

Generally, due to unpredictable and vulnerable features of the stock market, traditional statistical methods with single factor cannot explain the movements of stock prices well. In this paper, multi-factor model of quantitative stock selection is applied to predict stock price index to perceive the fluctuation of stock in multiple aspects and improve the accuracy of prediction. This study also combines machine learning measurements with historical data fitting effects. The data used for analysis is all collected from Tushare website, and the stock data are from the 2015 to 2022 data of Shanghai Pudong Development Bank and China Merchants Bank, and the selected factors include price-to-earnings ratio, price-to-book ratio, volume ratio, total equity and so forth. Ordinary least square linear regression and random forest nonlinear approaches are utilized to predict the stock price. According to the analysis, the accuracy of random forest is higher than OLS in stock price prediction. Among all the factors, the opening price, the highest price and the lowest price have relatively large impacts on the closing price of the next day. However, when selecting different factors, stocks and train-testing periods, the obtained regression models are different. Therefore, the determination of the correlation coefficient is not invariable and needs to be analyzed on a case-by-case basis. Overall, these results shed light on the superiority of machine learning model and the significant contribution of some factors on stock price prediction.