Presented By: Richa Handa Asst. Professor
Jan 23, 2017
Presented By:Richa HandaAsst. Professor
The stock market is a complex and dynamic system with noisy, non-stationary and chaotic data series.
Prediction of a financial market is more challenging due to chaos and uncertainty of the system. Soft computing techniques are progressively gaining presence in the financial world.
This paper describes the application of Artificial Neural Network (ANN) for the prediction of Stock Market using some technical indicators..
A new model is proposes of ANN for feature Extraction and selection to get more accurate prediction of stock exchange market.
In this research work a framework is designed for an optimal stock data prediction to develop an intelligent decision support system.
This developed system remove the non linearity that exist in financial time series data using some feature extraction and selection.
These extracted features is apply to model of ANN and data mining techniques to get the accurate prediction of stock price.
EBPN
ANN Techniques
SupervisedLearning
UnsupervisedLearning
KSOM
RBFN
Radial Basis Function (RBF) Neural Network:Radial basis functions are powerful techniques for interpolation in multidimensional space. A RBF is a function which has built into a distance criterion with respect to a center.
Error Back Propagation Network (EBPN): It is a supervised learning method, and is a generalization of the
delta rule. It requires a dataset of the desired output for many inputs, making up the training set. It is most useful for feed-forward networks
Feature extraction method is transformative: that is we are applying transformation to our data to project it into new feature space with lower dimension.
One of the essential features of data mining is feature selection, this technique is mostly based on the machine learning for selection set of feature for improving the efficiency of the prediction. Feature selection techniques to automatically discover the best features and it helps to solve the problems of having too much data.
The data used in this study consist of BSE30 and BSE100 data collected from the historical data available on the website yahoo finance.
This dataset encompasses five years data. The collected data is Non linear by nature, so preprocessing technique has been done to make the data smoother. For preprocessing of data some technical indicators are used suggested by some researchers.
1. Exponential Moving Average(EMA)2. Moving Average Convergence-Divergence(MACD)3. Relative Strength Index(RSI)4. Stochastic Oscillator5. Rate of Change(ROC)6. Money Flow Index(MFI)7. William %R8. Accumulation Distribution Line(A/D)9. On Balance Volume(OBV)10. Chaikin Oscillator(CHO)11. Average True Range12. Average Directional Index(ADX)13. Commodity Channel Index(CCI)14. Chaikin Money Flow(CMF)15. Percentage Price Oscillator(PPO)16. Force Index(FI)
In this approach the Bombay Stock Exchange(BSE) data are collected including as opening price, closing price, lowest price, highest price and volume.
At the second stage, variables that had less significant ability were removed and Feature extraction and selection will be done.
Technical Indicators
Data Smoothing Feature Extraction
ANN Model
Stock Data
Analyzed Data
Feature Selection
Ranking
Stock Data
Feature Extraction
Normalize Data
Partition Data
Feature Selection
Analyze DataANN Model
MAE MAPE RMSE
No Of FEATURES MAPE RMSE MAE
TRAINING
16 5.5138 0.03578 0.026
13 6.0853 0.0357 0.026
11 6.1105 0.0357 0.028
10 5.979 0.0344 0.025
9 5.5307 0.0342 0.025
8 5.5453 0.03428 0.025
7 5.4846 0.343 0.025
Best set of technical indicators will be extracted through optimization techniques.
The empirical result show that feature extraction and selection play a crucial role in term of robustness and efficiency of ANN model.