Neural network technology for pattern recognition, stock prediction and market forecasting


Pattern recognition

DCT-ANN Face Identification

Wavelet-ANN Face Recognition

Text-Independent Speaker Recognition based on ANN

Assembler-based Neural Network Simulator

Facial Expression Recognition System

Iris Recognition Based on Neural Networks

Neural Networks Based Signature Recognition

Eye Detection Based Facial Expression Recognition

Gait Recognition System

Leaf Recognition System

Optical Character Recognition

Neural Network Fingerprint Recognition

Keystroke Recognition

EEG Recognition

Neural Network Speech Recognition

Image processing

Image Compression With Neural Networks

Stock Market Forecasting

Neural Network Forecasting

External resources

Advanced Source Code .Com

Genetic Algorithms .It

Face Recognition .It

Iris Recognition .It

Stock Market Forecasting based on Neural Networks and Wavelet Decomposition

Download now Matlab source code
Requirements: Matlab, Matlab Image Processing Toolbox, Matlab Neural Network Toolbox, Matlab Wavelet Toolbox.

We have developed an efficient tool for intraday stock market forecasting based on Neural Networks and Wavelet Decomposition. This software has been tested on real data obtaining excellent results. SMF Tool gives Buy/Sell signals with a high degree of accuracy. SMF accepts, as input, a sequence of given length N. The system can determine if at least one of future prices - within an observation window of fixed length M - will be higher or lower than current price. SMF package has been tested with Italian Futures over a period of 3 years, more than 600 days of effective trading. Training data and testing data have been randomly selected from this data set, without any overlapping. Stock market data have been downloaded at these data are uniformly sampled each minute.

Why Wavelets
Wavelets can localize data in time-scale space. At high scales (shorter time intervals), the wavelet has a small time support and is thus, better able to focus on short lived, strong transients like discontinuities, ruptures and singularities. At low scales (longer time intervals), the wavelet's time support is large, making it suited for identifying long periodic features. Wavelets have a intuitive way of characterizing the physical properties of the data. At low scales, the wavelet characterizes the data's coarse structure; its long-run trend and pattern. By gradually increasing the scale, the wavelet begins to reveal more and more of the data's details, zooming in on its behavior at a point in time. Wavelet analysis is the analysis of change. A wavelet coefficient measures the amount of information that is gained by increasing the frequency at which the data is sampled, or what needs to be added to the data in order for it to look like it had been measured more frequently. For instance, if a stock price does not change during the course of a week, the wavelet coefficients from the daily scale are all zero during that week. Wavelet coefficient that are non-zero at high scales typically characterize the noise inherent in the data. Only those wavelets at very fine scales will try to follow the noise, whereas those wavelets at coarser scales are unable to pick up the high frequency nature of the noise.

Why Neural Networks
Since the early 90's when the first practically usable types emerged, artificial neural networks (ANNs) have rapidly grown in popularity. They are artificial intelligence adaptive software systems that have been inspired by how biological neural networks work. Their use comes in because they can learn to detect complex patterns in data. In mathematical terms, they are universal non-linear function approximators meaning that given the right data and configured correctly, they can capture and model any input-output relationships. This not only removes the need for human interpretation of charts or the series of rules for generating entry/exit signals but also provides a bridge to fundamental analysis as that type of data can be used as input. In addition, as ANNs are essentially non-linear statistical models, their accuracy and prediction capabilities can be both mathematically and empirically tested. In various studies neural networks used for generating trading signals have significantly outperformed buy-hold strategies as well as traditional linear technical analysis methods. While the advanced mathematical nature of such adaptive systems have kept neural networks for financial analysis mostly within academic research circles, in recent years more user friendly neural network software has made the technology more accessible to traders.

Index Terms: Matlab source code, price, neural networks, stock market prediction, neural network, wavelet, decomposition, wavelets, stock market forecasting, data, model, business, financial, analysis, target, marketing, optimization.

Release 1.0 Date 2007.03.10
Major features:

Neural Networks . It Luigi Rosa mobile +39 3207214179