Is It Possible to Earn Abnormal Return in an Inefficient Market? An Approach Based on Machine Learning in Stock Trading

Khoa, Bùi Thành; Huynh, Tran Trong

doi:10.1155/2021/2917577

Cited by 13 publications

(5 citation statements)

References 60 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are several reasons to explain this phenomenon. First, Pham and Tatiana (2020) and Bui and Tran (2021) conclude that Vietnam's stock exchange is inefficient in weak form with inadequate access to reliable information. Therefore, investors respond more slowly to information and often adhere to market consensus.…”

Section: Resultsmentioning

confidence: 99%

Herding behavior before and after COVID-19 pandemic: evidence from the Vietnam stock market

Nguyen

2023

JES

View full text Add to dashboard Cite

PurposeThe priority of this study is to contribute to the literature by examining herding behavior at different periods of the COVID-19 pandemic. Furthermore, this study aims to investigate the herding behavior conditioned on market liquidity and information demand.Design/methodology/approachThis study investigates herding behavior in Vietnam's stock exchanges (Ha Noi Stock Exchange and Ho Chi Minh Stock Exchange) on a sample of daily stock closing prices of 425 firms from 2018 to the first half of 2022.FindingsThe research confirms the existence of herding behavior not only for the whole but also during and post-COVID periods. These results are robust in both bull and bear markets, further confirming the influence of COVID-19 on herding in Vietnamese background. Moreover, when the authors condition exogenous factors for each period, the herding tendency is more evident at the medium market liquidity level than at high and low levels. Besides, the pandemic causes herding behavior of investors with low and medium information demand.Research limitations/implicationsThese findings imply some recommendations that facilitate investors, policymakers and researchers in the context of the COVID-19 crisis.Originality/valueThe study contributes to the herding literature by examining herd behavior during the post-COVID period, suggesting the long-term impact of the health crisis. Furthermore, the research provides new evidence of herding behavior conditioned on market liquidity and information demand during different COVID sub-periods.

show abstract

Section: Resultsmentioning

confidence: 99%

Herding behavior before and after COVID-19 pandemic: evidence from the Vietnam stock market

Nguyen

2023

JES

View full text Add to dashboard Cite

show abstract

“…The buying and selling system developed using the NPMM labeling method can work well if the evaluation is good. The author evaluates the trading system using learning data figures, win ratio (Wr), payout ratio (Pr), and profit factor (Pf), as the graphs above show [4][5][6].…”

Section: Fig 2 Stock Market System Flow Chartmentioning

confidence: 99%

“…The rolling window approach refers to analyzing the stability of parameters in the model. The rolling window technique can accurately predict the parameters of a given model developed by machine learning algorithms [4][5][6].…”

Section: Introductionmentioning

confidence: 99%

Application of Machine Learning Algorithms in the Stock Market Analysis

Li¹

2023

HBEM

View full text Add to dashboard Cite

With the development of deep learning and machine learning, more new methods have been produced in the economic and financial fields. When talking about machines, one thing that comes to people’s minds is what they can do with machines to solve problems that need machines. The people in the stock market always want to find ways to forecast the stock trend, the pattern of stock, and the stock value. Before the development of machine learning algorithms, stock market predictions could be made in limited ways, and those methods usually did not produce accurate predictions. However, machine learning algorithms changed the phenomenon and offered people novel ways to analyze the stock market. This paper will discuss three research in which authors have implemented machine learning algorithms into stock market analyses. From analyzing the research, this paper tries to investigate the extent of applying machine learning algorithms in the stock market and how the algorithms have helped investors make improvements in stock market analysis.

show abstract

“…We have 𝑤𝑥 𝑇 + 𝑏 = 0, where w and b are the coefficients. The coefficients w and b should be chosen such that 𝑤𝑥 𝑇 + 𝑏 ≥ 1 if vnic, vnipc related to the market portfolio are statistically significant at 0.05 [24]. The negative sign of the vnic estimator implies that the greater the market volatility, the lower the probability of an upward movement of the closing price is expected.…”

Section: Support Vector Machine (Svm)mentioning

confidence: 99%

Forecasting stock price movement direction by machine learning algorithm

Khoa

Huynh

2022

IJECE

Self Cite

View full text Add to dashboard Cite

<p><span lang="EN-US">Forecasting stock price movement direction (SPMD) is an essential issue for short-term investors and a hot topic for researchers. It is a real challenge concerning the efficient market hypothesis that historical data would not be helpful in forecasting because it is already reflected in prices. Some commonly-used classical methods are based on statistics and econometric models. However, forecasting becomes more complicated when the variables in the model are all nonstationary, and the relationships between the variables are sometimes very weak or simultaneous. The continuous development of powerful algorithms features in machine learning and artificial intelligence has opened a promising new direction. This study compares the predictive ability of three forecasting models, including <a name="_Hlk106797328"></a>support vector machine (SVM), artificial neural networks (ANN), and logistic regression. The data used is those of the stocks in the VN30 basket with a holding period of one day. With the rolling window method, this study got a highly predictive SVM with an average accuracy of 92.48%.</span></p>

show abstract

Is It Possible to Earn Abnormal Return in an Inefficient Market? An Approach Based on Machine Learning in Stock Trading

Cited by 13 publications

References 60 publications

Herding behavior before and after COVID-19 pandemic: evidence from the Vietnam stock market

Herding behavior before and after COVID-19 pandemic: evidence from the Vietnam stock market

Application of Machine Learning Algorithms in the Stock Market Analysis

Forecasting stock price movement direction by machine learning algorithm

Contact Info

Product

Resources

About