Stock Market Prediction of NIFTY 50 Index Applying Machine Learning Techniques

Figures & data

Figure 1. Recurrent neural network cell structure (Ozbayoglu, Gudelek, and Sezer 2020).

Figure 2. Long short-term memory cells (Ozbayoglu, Gudelek, and Sezer 2020).

Figure 3. Convolutional neural network structure (Albelwi and Ausif 2017).

Table 1. A comparative study of state-of-the-art of deep learning model for stock market prediction.

	Scope			Prediction Method					DS	Results
References	Market	Feature/s	Prediction/s	NN	RNN	LSTM	CNN	DRL	MASS	PM	Values
(AbdelKawy, Abdelmoez, and Shoukry 2021)	USA SM	OHLCV	stock			✓		✓	✓	APR	0.73
										AMDD	0.217
(Lee et al. 2020)	USA SM	close	stock				✓	✓	✓	Return	6.123
			price							Sharpe ratio	2.247
(Zhang, Zohren, and Roberts 2020)	SSE	CSI 300	stock					✓		SD	1.9586
		stock	return							Accuracy	0.038
(Kim and Won 2018)	korea	KOSPI 200	stock	✓		✓				MAE	0.46%
	stock	index	price							MSE	7.5%
	exchange									HMAE	29%
										HMSE	45.7%
(Hiransha et al. 2018)	NYSE	Close	stock		✓	✓	✓			MAPE	4.05
	NSE	Price	Price
(Chen and He 2018)	Chinese SM	OHLCV	stock				✓			Accuracy	75,2%
			Price							F1-score	73%
(Gurjar et al. 2018)	NSE	NIFTY 50	stock	✓						SD	-
		index	trend							MA	-
(Jain, Gupta, and Moghe 2018)	NSE	NIFTY 50	Market	✓						CE	6.13%
		index	direction
(Hu et al. 2018)	Chinese SM	Price	Stock	✓	✓					Accuracy	0.478
		Volume	Trend							ARR	0.611
(Fabbri and Moro 2018)	USA SM	DJIA	Stock		✓	✓				Accuracy	83%
		index	time series
(Minh et al. 2018)	USA SM	S&P 500	stock			✓				Accuracy	66.32%
		Index	trends							Precision	72.1%
										Recall	77.32%
(Liu, Zhang, and Ma 2017)	Chinese SM	Chinese	stock			✓	✓			ARR	0.309
		stocks	return
(Chong, Han, and Park 2017)	South Korea SM	KOSPI	stock	✓						RMSE	0.8214
		index	return							NMSE	0.9621
										MAE	0.5838
(Vargas, De Lima, and Evsukoff 2017)	USA SM	S&P 500	Market		✓		✓			Accuracy	64,21%
		Index	direction
(Roondiwala, Patel, and Varma 2017)	NSE	NIFTY 50	Stock		✓	✓				MSE	-
		Index	price
(Heaton, Polson, and Witte 2017)	NASDAQ	IBB	Market	✓						RMSE	2.36
		index	direction
(Selvin et al. 2017)	NSE	NIFTY IT	Stock		✓	✓	✓			RMSE	2.36
		index	price
(Huynh, Dang, and Duong 2017)	USA SM	S&P 500	Stock			✓				Accuracy	59.98%
		index	Price
(Bao, Yue, and Rao 2017)	Stock	S&P 500	stock		✓	✓
	markets	DJIA, HSI	price							MAPE	0.011
		Nifty 50								R	0.949
		CSI 300								Theil U	0.007
		Nikkeiindices
(Singh and Srivastava 2017)	NASDAQ	Close	stock		✓					Accuracy	43.4%
		price	Price							RMSE	0.007
										MAPE	0.08

Table 2. Loss of neural network models.

Loss
Activation Function	RNN	LSTM	CNN
ReLu	0.6150	0.4705	0.6291
Sigmoid	0.6170	0.5203	0.6114
Softmax	0.6510	0.4828	0.7258

Figure 4. Validation Loss of LSTM model using Relu activation function.

Table 3. Error value for LSTM training set.

Features	Epoc.	RMSE	MSE	MAE	R^2
O – C	250	0.012	0.025	0.035	0.31
O – C	500	0.035	0.048	0.027	0.36
H - L – C	250	0.015	0.016	0.061	0.51
H - L- C	500	0.011	0.026	0.054	0.21
H - L- O- C	250	0.053	0.049	0.073	0.67
H - L - O – C	500	0.004	0.008	0.006	0.54
H - L - O - C- V	500	0.005	0.009	0.008	0.37

Figure 5. Real value vs Predicted value for NIFTY 50 using RNN .

Table 4. Error value for LSTM testing set.

Features	Epoc.	RMSE	MSE	MAE	R^2
O – C	250	0.044	0.062	0.081	0.528
O – C	500	0.055	0.051	0.067	0.341
H - L – C	250	0.037	0.023	0.032	0.516
H - L – C	500	0.063	0.053	0.034	0.201
H - L - O – C	250	0.082	0.057	0.063	0.537
H - L - O – C	500	0.002	0.009	0.006	0.482
H - L - O - C – V	500	0.004	0.005	0.002	0.247

Table 5. Error value for RNN training set.

Features	Epoc.	RMSE	MSE	MAE	R^2
O – C	250	0.014	0.085	0.072	0.401
O – Close	500	0.035	0.073	0.015	0.351
H - L – C	250	0.063	0.092	0.048	0.578
H - L – C	500	0.028	0.052	0.038	0.427
H - L - O – C	250	0.069	0.064	0.083	0.563
H - L - O – C	500	0.007	0.004	0.009	0.358
H - L - O - C – V	500	0.006	0.005	0.007	0.497

Table 6. Error value for RNN testing set.

Features	Epoc.	RMSE	MSE	MAE	R^2
O – C	250	0.029	0.058	0.037	0.485
O – C	500	0.045	0.057	0.057	0.274
H - L – C	250	0.047	0.059	0.045	0.810
H - L – C	500	0.079	0.047	0.048	0.389
H - L - O – C	250	0.047	0.037	0.036	0.584
H - L - O – C	500	0.007	0.003	0.007	0.804
H - L - O - C – V	500	0.007	0.006	0.004	0.638

Table 7. Error value for CNN training set.

Features	Epoc.	RMSE	MSE	MAE	R^2
O – C	250	0.013	0.017	0.017	0.359
O – C	500	0.018	0.026	0.047	0.642
H - L – C	250	0.018	0.017	0.025	0.361
H - L – C	500	0.032	0.024	0.034	0.675
H - L - O – C	250	0.039	0.024	0.074	0.819
H - L - O – C	500	0.016	0.015	0.014	0.520
H - L - O - C – V	500	0.017	0.016	0.015	0.371

Figure 6. Real value vs predicted value for NIFTY 50 using LSTM .

Figure 7. Prediction of close price using RNN.

Figure 8. Prediction of close price using CNN.

Figure 9. Prediction of close price using LSTM.

Ozbayoglu, A. M., M. U. Gudelek, and O. B. Sezer. 2020. Deep learning for financial applications : A survey. Applied Soft Computing 93:106384. doi:10.1016/j.asoc.2020.106384.

 Web of Science ®Google Scholar

Albelwi, S., and M. Ausif. 2017. A framework for designing the architectures of deep convolutional neural networks. Entropy 19 (6):242. doi:10.3390/e19060242.

 Web of Science ®Google Scholar

AbdelKawy, R., W. M. Abdelmoez, and A. Shoukry. 2021. A synchronous deep reinforcement learning model for automated multi-stock trading. Progress in Artificial Intelligence 10 (1):83–3123. doi:10.1007/s13748-020-00225-z.

 Web of Science ®Google Scholar

Lee, J., R. Kim, S.-W. Yi, and J. Kang. 2020. Maps: Multi-agent reinforcement learning-based portfolio management system. arXiv preprint arXiv:200705402.

 Google Scholar

Zhang, Z., S. Zohren, and S. Roberts. 2020. Deep reinforcement learning for trading. The Journal of Financial Data Science 2 (2):25–40. doi:10.3905/jfds.2020.1.030.

 Google Scholar

Kim, H. Y., and C. H. Won. 2018. Forecasting the volatility of stock price index: A hybrid model integrating lstm with multiple garch-type models. Expert Systems with Applications 103:25–37. doi:10.1016/j.eswa.2018.03.002.

 Web of Science ®Google Scholar

Hiransha, M., E. A. Gopalakrishnan, V. K. Menon, and K. Soman. 2018. Nse stock market prediction using deep-learning models. Procedia Computer Science 132:1351–62. doi:10.1016/j.procs.2018.05.050.

 Google Scholar

Chen, S., and H. He (2018). Stock prediction using convolutional neural network. In IOP Conference series: materials science and engineering, Volume 435, pp. 012026. IOP Publishing.

 Google Scholar

Gurjar, M., P. Naik, G. Mujumdar, and T. Vaidya. 2018. Stock market prediction using ann. International Research Journal of Engineering and Technology 5 (3):2758–61.

 Google Scholar

Jain, S., R. Gupta, and A. A. Moghe (2018). Stock price prediction on daily stock data using deep neural networks. In 2018 International conference on advanced computation and telecommunication (ICACAT), pp. 1–13. IEEE.

 Google Scholar

Hu, Z., W. Liu, J. Bian, X. Liu, and T.-Y. Liu (2018). Listening to chaotic whispers: A deep learning framework for news-oriented stock trend prediction. In Proceedings of the eleventh ACM international conference on web search and data mining, pp. 261–69.

 Google Scholar

Fabbri, M., and G. Moro. 2018. Dow jones trading with deep learning: The unreasonable effectiveness of recurrent neural networks. Data 142–53.

 Google Scholar

Minh, D. L., A. Sadeghi-Niaraki, H. D. Huy, K. Min, and H. Moon. 2018. Deep learning approach for short-term stock trends prediction based on two-stream gated recurrent unit network. IEEE Access 6:55392–404. doi:10.1109/ACCESS.2018.2868970.

 Web of Science ®Google Scholar

Liu, S., C. Zhang, and J. Ma (2017). Cnn-Lstm neural network model for quantitative strategy analysis in stock markets. In international conference on neural information processing, pp. 198–206. Springer.

 Google Scholar

Chong, E., C. Han, and F. C. Park. 2017. Deep learning networks for stock market analysis and prediction: Methodology, data representations, and case studies. Expert Systems with Applications 83:187–205. doi:10.1016/j.eswa.2017.04.030.

 Web of Science ®Google Scholar

Vargas, M. R., B. S. De Lima, and A. G. Evsukoff (2017). Deep learning for stock market prediction from financial news articles. In 2017 IEEE international conference on computational intelligence and virtual environments for measurement systems and applications (CIVEMSA), pp. 60–65. IEEE.

 Google Scholar

Roondiwala, M., H. Patel, and S. Varma. 2017. Predicting stock prices using lstm. International Journal of Science and Research (IJSR) 6 (4):1754–56.

 Google Scholar

Heaton, J. B., N. G. Polson, and J. H. Witte. 2017. Deep learning for finance: Deep portfolios. Applied Stochastic Models in Business and Industry 33 (1):3–12. doi:10.1002/asmb.2209.

 Web of Science ®Google Scholar

Selvin, S., R. Vinayakumar, E. Gopalakrishnan, V. K. Menon, and K. Soman (2017). Stock price prediction using lstm, rnn and cnn-sliding window model. In 2017 International Conference on Advances in Computing, Communications and Informatics (ICACCI), pp. 1643–47. IEEE.

 Google Scholar

Huynh, H. D., L. M. Dang, and D. Duong (2017). A new model for stock price movements prediction using deep neural network. In Proceedings of the Eighth International Symposium on Information and Communication Technology, pp. 57–62.

 Google Scholar

Bao, W., J. Yue, and Y. Rao. 2017. A deep learning framework for financial time series using stacked autoencoders and long-short term memory. PloS One 12 (7):e0180944. doi:10.1371/journal.pone.0180944.

 PubMed Web of Science ®Google Scholar

Singh, R., and S. Srivastava. 2017. Stock prediction using deep learning. Multimedia Tools and Applications 76 (18):18569–84. doi:10.1007/s11042-016-4159-7.

 Web of Science ®Google Scholar