An Empirical Research and Comprehensive Analysis of Stock Market Prediction using Machine Learning and Deep Learning techniques

Singh, Aryendra; Gupta, Priyanshi; Thakur, Narina

doi:10.1088/1757-899x/1022/1/012098

Cited by 19 publications

(5 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finding reliable patterns to forecast the values of a given system is a task for which machine learning algorithms are increasingly applied; see, for example, [55][56][57][58][59][60][61]. This works well when the task is set in a static environment, such as optical character recognition or the classification of pictures with animals or human faces.…”

Section: Causality Connection Strength and Explainability In Machine ...mentioning

confidence: 99%

Quantifying and Interpreting Connection Strength in Macro- and Microscopic Systems: Lessons from Bell’s Approach

Gallus

Błasiak

Pothos

2022

Entropy

View full text Add to dashboard Cite

Bell inequalities were created with the goal of improving the understanding of foundational questions in quantum mechanics. To this end, they are typically applied to measurement results generated from entangled systems of particles. They can, however, also be used as a statistical tool for macroscopic systems, where they can describe the connection strength between two components of a system under a causal model. We show that, in principle, data from macroscopic observations analyzed with Bell’ s approach can invalidate certain causal models. To illustrate this use, we describe a macroscopic game setting, without a quantum mechanical measurement process, and analyze it using the framework of Bell experiments. In the macroscopic game, violations of the inequalities can be created by cheating with classically defined strategies. In the physical context, the meaning of violations is less clear and is still vigorously debated. We discuss two measures for optimal strategies to generate a given statistic that violates the inequalities. We show their mathematical equivalence and how they can be computed from CHSH-quantities alone, if non-signaling applies. As a macroscopic example from the financial world, we show how the unfair use of insider knowledge could be picked up using Bell statistics. Finally, in the discussion of realist interpretations of quantum mechanical Bell experiments, cheating strategies are often expressed through the ideas of free choice and locality. In this regard, violations of free choice and locality can be interpreted as two sides of the same coin, which underscores the view that the meaning these terms are given in Bell’s approach should not be confused with their everyday use. In general, we conclude that Bell’s approach also carries lessons for understanding macroscopic systems of which the connectedness conforms to different causal structures.

show abstract

Section: Causality Connection Strength and Explainability In Machine ...mentioning

confidence: 99%

Quantifying and Interpreting Connection Strength in Macro- and Microscopic Systems: Lessons from Bell’s Approach

Gallus

Błasiak

Pothos

2022

Entropy

View full text Add to dashboard Cite

show abstract

“…In [23], authors have proposed an LSTM model for stock market prediction with computation of Linear Regression. They have used K-NN classifier for classification of dataset, computed the moving average of the stock TITAN and NIFTY50.…”

Section: Review Of Literaturementioning

confidence: 99%

Untitled

2021

CIML

View full text Add to dashboard Cite

The elementary goal of this paper is to predict the best model for estimation of stock market. Machine Learning is a blooming field in computer science that has contributed to many predictions and analysis-based algorithm in Financial and economical field. Some of the algorithms used for predictions are Random Forest (RF), Support vector machine (SVM), Long-Short Term Memory (LSTM), Artificial Neural Networks (ANN). Random Forest is an ensemble supervised learning algorithm for classification problems with high accuracy factor. ANN has matured to a great extend over the past years. With the advent of high-performance computing ANN has assumed tremendous significance and huge application potentials in recent years. The innovation of ANN technology mimics the large interconnections and networking that exists between the nerve cells to process complex task. The paper has presented ANN and RF model for stock price estimation based on historical data and computed the future price, with comparative result analysis of their performance. Further, a candlestick model is designed of the stock to show the variation in price of stock over a stipulated period of time

show abstract

“…From this study, 97% accuracy was obtained using historical stock price data in 5 companies such as GOOGLE, APPLE, MICOROST, AMAZON and VIX for the last 10 years. Another study using the KNN and K-means algorithms in the process of predicting stock prices, for data collection companies are grouped into 3 parts, namely companies with small capitalization (Small Cap), medium (Mid cap Company), and large capital (Big Cap) with Rupee currency, This study shows a success of 0.013, which means the algorithm is quite accurate and the resulting Beta is 0.609, which means for the last three months [10]. The application of the Hidden Markov Model (HMM) method to predict the stock prices of Apple, Google, and Facebook.…”

Section: Introductionmentioning

confidence: 98%

Application of Triple Exponential Smoothing Method to Predict LQ45 Saham Stock Price

Nurdin¹

2022

CoreIT

View full text Add to dashboard Cite

The capital market is one of the investment models that is currently growing so rapidly because there are more and more digital-based investment platforms that can be accessed using mobile smartphones. The amount of interest in investing makes many people who experience losses due to not understanding the investment risks. For this reason, it is necessary to have the ability to analyze technically based on historical data. The object of this research is LQ45 shares in three companies, Indofood Sukses Makmur Tbk (INDF), Unilever Indonesia Tbk (UNVR), and Aneka Tambang Tbk (ANTM). The method used in this research is the Triple Exponential Smoothing method which is a prediction method that utilizes the statistical analysis method. The variables used in this study are historical prices ranging from Open, High, Low, and Close prices. The stages used are the collection of 125 historical data, where the data is taken through the Google Finance financial database. Then the Triple Exponential Smoothing calculation process is carried out, the data is stored in the database and presented in the form of graphs and tables. By using the parameter values = 0.13 and = 0.87 in the end it produces a Mean margin error level of Open price -0.10681%, High price -1.1156%, Low price 1.4616%, and Close price -0.2504%. The results of the study mean the margin of error is between -0.1% to 1%. The application of Triple Exponential Smoothing can be applied to predict stock prices. This research is to help investors analyze stock price movements.

show abstract

An Empirical Research and Comprehensive Analysis of Stock Market Prediction using Machine Learning and Deep Learning techniques

Cited by 19 publications

References 22 publications

Quantifying and Interpreting Connection Strength in Macro- and Microscopic Systems: Lessons from Bell’s Approach

Quantifying and Interpreting Connection Strength in Macro- and Microscopic Systems: Lessons from Bell’s Approach

Untitled

Application of Triple Exponential Smoothing Method to Predict LQ45 Saham Stock Price

Contact Info

Product

Resources

About