Performance analysis of LSTM model on equity domain data

##plugins.themes.academic_pro.article.main##

Lalit Agrawal
Dattatraya Adane

Abstract

Predicting the stock market trend is a difficult problem because of its dynamic nature and other external factors
affecting it. This research work is done to assess the trend in stock prices and to check whether using different
trading prices like open, high and low instead of using just the closing price gives more accurate results. For this
purpose, we have used the long short term memory (LSTM) model and trained this model by using historic stock
prices data. We created three different data frames from the historic stock price data (from the year 2000-2019)
such that the first data frame contains median stock price calculated by finding the median of all the trading prices
for each row of data. Similarly, the second data frame contains max stock prices among all the trading prices and
the third data frame contains min stock price among the trading price. Then three different LSTM models are
created and trained using these three data frames. We then test these three models by using them to predict the
stock prices for the year 2020 and plot a graph with traces for the actual stock prices of the year 2020 and predict
stock prices by these three models. It was observed from the plotted graph that the traces for the actual and
predicted stock prices were very close to each other which means that the models are giving satisfactory results.

##plugins.themes.academic_pro.article.details##

How to Cite
Agrawal, L., & Adane, D. (2021). Performance analysis of LSTM model on equity domain data. International Journal of Next-Generation Computing, 12(5). https://doi.org/10.47164/ijngc.v12i5.437

References

  1. Brownlee, J. 2017. Long short-term memory networks with python: develop sequence prediction models with deep learning. Machine Learning Mastery.
  2. Education, I. C. 2020. Recurrent neural networks. source: https://www. ibm.com/cloud/learn/recurrent-neural-networks.
  3. Ghosh, A., Bose, S., Maji, G., Debnath, N., and Sen, S. 2019. Stock price prediction using lstm on indian share market. In Proceedings of 32nd international conference on. Vol. 63. 101–110.
  4. Hiransha, M., Gopalakrishnan, E. A., Menon, V. K., and Soman, K. 2018. Nse stock market prediction using deep-learning models. Procedia computer science 132, 1351–1362.
  5. Moghar, A. and Hamiche, M. 2020. Stock market prediction using lstm recurrent neural network. Procedia Computer Science 170, 1168–1173.
  6. Moghar, A. and Hamiche, M. 2020. Stock market prediction using lstm recurrent neural network. Procedia Computer Science 170, 1168–1173.