Predicting Support and Resistance Indicators for Stock Market with Fibonacci Sequence in Long Short-Term Memory

Predicting Support and Resistance Indicators for Stock Market with Fibonacci Sequence in Long Short-Term Memory

Journal of Computer Science Original Research Paper Predicting Support and Resistance Indicators for Stock Market with Fibonacci Sequence in Long Short-Term Memory 1Thambusamy Velmurugan and 2Thiruvalluvan Indhumathy 1Department of Computer Applications, D.G. Vaishnav College, Chennai, India 2Department of Computer Applications, D. G. Vaishnav College, Chennai-600106, Tamil Nadu India, India Article history Abstract: Predictive data analytics is a branch of data analytics where Received: 30-08-2020 models are designed that effectively interpret; anticipate outcomes by Revised: 26-10-2020 analyzing present data to make predictions about future. One of the major Accepted: 28-10-2020 attributes for prediction is time and there have been a considerable number of time series analytical models that are used for forecasting. Long Short- Corresponding Author: Thambusamy Velmurugan Term Memory (LSTM) is a deep learning model which is used as a time Department of Computer series model and this research work had made an attempt to apply Science, D.G. Vaishnav Fibonacci sequence for retracement of the support and resistance levels, one College, India of the commonly used trend indicators and used LSTM for predicting those Email: [email protected] levels. These levels help identify the uptrend or downtrend that decide the buying or holding of shares. For this purpose, datasets from financial sector was taken and split into 80% of training data and 20% of testing data for the analysis. The source of these datasets is Kaggle and each of these dataset has a total of 5021 instances from January 2000 till February 2020. Scaling of data was done, retracement of values using three Fibonacci percentages along with the starting and ending level was applied using the LSTM network. For measuring the accuracy of the model, the Root-Mean-Square Error (RSME) metric was used. The comparison between the error score with the lowest value of the dependent variable determines the level of accuracy which is expected to be less in LSTM model that is to be conformed as a test case. Keywords: Support Value, Resistance Value, Fibonacci Retracement, Long Short-Term Memory, Root-Mean-Square Error Metric Introduction usually taken at successive equally spaced points in time. This sequence is plotted via line charts usually as With rising number of data generated data analytics independent variable against any other attribute as has a deep significance in areas that needs insights into dependent variable that is specific to the application. future. These insights set trends, achieve goals, reduce Time series analytical (Wei, 2006) methods are used for costs, improve efficiency, increase productivity and extracting the features and deducing essential statistical sharpen the performance. Raw data with varied data while time series forecasting methods and models is attributes, high dimension and multiple sets are searched adopted for prediction based on historical data. and properties from them are extracted through mining Time series analysis and forecasting are used in techniques to form a structured set. Statistical tools are industries, banking and financial services, cross marketing, applied to these sets and to derive inferences from these medical diagnosis, marketing and Sales forecasting, security formed structures, machine learning and deep learning issues, collection analysis, risk management, fraud algorithms are put to use. Out of the varied attributes, detection, underwriting, statistics, signal processing, pattern time is considered as a significant attribute in data recognition, econometrics, mathematics, finance, weather analytics that is used in analyzing, predicting and forecasting, earthquake prediction, electroencephalography, forecasting - yet it has often been neglected. In order to control engineering, astronomy, communications make use of this attribute it is taken in the form of series engineering and largely in any domain of applied science © 2020 Thambusamy Velmurugan and Thiruvalluvan Indhumathy. This open access article is distributed under a Creative Commons Attribution (CC-BY) 3.0 license. Thambusamy Velmurugan and Thiruvalluvan Indhumathy / Journal of Computer Science 2020, 16 (10): 1428.1438 DOI: 10.3844/jcssp.2020.1428.1438 and engineering which involves temporal measurements. Related Work For all the above mentioned time series applications, deep learning models (Heaton et al., 2016) are used for There are many papers dedicated to emphasize the time series analysis (Gamboa, 2017) and time series amazing characteristics of Fibonacci series, expose the forecasting (Qiu et al., 2014), it could also be combined hidden aesthetics and its usage in various applications. with other techniques (Moews et al., 2019) and these Few papers that are related to the application of the models are proven to be effective than the other sequence in stock markets and the technical aspect of traditional methods. stock market have been discussed below. Recurrent Time series forecasting implemented by deep Neural Networks (RNN) and Long Short-Term Memory learning models is used to find the trends in share (LSTM) are considered to be models of importance in deep markets and there are many technical market indicators learning for prediction and are applied in various domains where a framework can be formed from which the where it is considered to be difficult to forecast. Here an investors can make two very important decisions-when attempt is made to expose few papers that discuss the to buy stocks and when to sell stocks. Support and application of the model Long Short-Term Memory in resistance levels are one of the technical indicators prediction of various attributes in share markets. that help investors to trace the past movements of Osler (2000) have defined about support and share prices where these prices fluctuate with supply resistance which is a technical analysis calculated to and demand. Support is where it is considered that find out the exchange rates that are interrupted or there will be no further decline in prices. The break in reversed. In this article the predictive power of rising of prices is considered as Resistance. support and resistance indicators have been examined Identifying these levels is an essential component for with the data containing yen, mark and pound that was a successful technical analysis. relative to dollar. The support and resistance level is Most of the mathematicians and scientists accept taken as one of the criteria to provide information Fibonacci sequence (Scott and Marketos, 2014) and regarding foreign exchange market for indicating the Golden ratio (Grigas, 2013) as an amazing concept ups and downs in prices. The findings reveal that the because of the presence of its sequence and pattern in rate trends were affected by both support and objects and activities in nature. Nature uses this concept resistance level more often than that was actually to maintain balance and seem to derive predictable expected. All the six firms that were taken for study patterns from atoms and astronomical objects. It is had almost similar results that lasted for a week. aesthetically pleasing and its mathematical properties are Kumar (2014) proposes a model in machine learning applied in many domains like mathematics, computer that can be used to remove the subjectivity that arises in science, music and architecture to mention a few. This domains like share market where decisions are made concept is also used as a technical tool and from a from predictions and specify it objectively. The author stock analyst’s point of view Fibonacci sequence is takes point of sale, money flow index, relative strength used for following the price pattern and how the support and resistance levels are retraced. These help index and chaiken money flow as independent variables to know about the possible areas of trending of price and a series of Fibonacci retracement of waves as waves where a trader can analyze and make decisions dependent values. Only the important variables are to sell or hold back the stocks. extracted by Factor analysis and curtail the Fibonacci This paper examines how the Fibonacci sequence is series to cardinality by two because there are actually a set developed as a retracement to know the support and of values. The study concludes that the wave 2 predicts resistance levels in share prices. This is then fitted in better than the rest by the proposed model. Mohammad. LSTM network model for future prediction and the (Alalaya and Almahameed, 2018) have taken Amman stock accuracy is sought out. The organization of this paper is exchange Market prices as dataset and have analyzed the as follows. Section 2 examines the work carried out by application of Fibonacci retracement and Elloit waves other authors that is in compliance with Fibonacci theory that determine the trading strategy, the time to buy sequence as a tool for prediction in stock markets, in and sell stocks and shares. They conclude that the sequence addition a review about support and resistance is also is helpful and helps to take decision to buy or hold. presented. The papers that discusses about long short- Pang et al. (2020) propose the deep Long Short-Term term memory is also been assessed. Section 3 Memory neural network (LSTM) with embedded layer elaborates about long short-term memory and about and the long short-term memory neural network with the facts related to Fibonacci sequence. The automatic encoder to predict the stock market for the methodology is also discussed in detail here. The Shanghai A-shares composite index. Comparison experimental results are put forth in section 4 and the between LSTM and ELSTM was made and conclusion about the paper is in section 5. demonstrated that ELSTM model has higher accuracy 1429 Thambusamy Velmurugan and Thiruvalluvan Indhumathy / Journal of Computer Science 2020, 16 (10): 1428.1438 DOI: 10.3844/jcssp.2020.1428.1438 than the LSTM. Roondiwala et al. (2017) uses Materials and Methods Recurrent Neural Network (RNN) and Long Short- Term Memory to predict stock market indices because Predicting the next move for a financial dataset using forecasting has been a difficult task.

View Full Text

Details

  • File Type
    pdf
  • Upload Time
    -
  • Content Languages
    English
  • Upload User
    Anonymous/Not logged-in
  • File Pages
    11 Page
  • File Size
    -

Download

Channel Download Status
Express Download Enable

Copyright

We respect the copyrights and intellectual property rights of all users. All uploaded documents are either original works of the uploader or authorized works of the rightful owners.

  • Not to be reproduced or distributed without explicit permission.
  • Not used for commercial purposes outside of approved use cases.
  • Not used to infringe on the rights of the original creators.
  • If you believe any content infringes your copyright, please contact us immediately.

Support

For help with questions, suggestions, or problems, please contact us