A tool which provides short-term forecast for stock traders
Neural networks can discover patterns in data that humans might not notice and successfully predict the future trend. Addaptron Software has developed NNSTP-1, neural network computer tool, to help stock traders in predicting stock prices for short terms. Neural Network Stock Trend Predictor is an application which predicts future share prices or their percentage changes (can be chosen in settings menu) using Fuzzy Neural Network (FNN). It operates automatically when creating the FNN, training it, and mapping to classify a new input vector. Recommended forecasting horizon is 5-10 trading days. Input data are EOD weighted closing price of the stock and the volume traded of the stock for the period that is set by user (normally, 50-100 days). The software allows using an advanced forecasting. It is based on automatic back testing scan of different inputs periods to define the accuracy for each period. Then the final forecast is built on weighted averaging of all forecasts proportionally to the accuracy of each inputs period forecast. Neural Network Stock Trend Predictor has a user-friendly easy-to-use interface. The software is intended for traders with a basic knowledge in stock analysis. Simply How Neural Network Stock Trend Predictor WorksThe algorithm of using Neural Network Stock Trend Predictor is simple - select input data, estimate predictability, and build forecast. The input data are historical quotes of the stock selected by user for some period (automatic downloading). Downloaded CSV-files from the Internet are stored in INPUT subfolder and can be used for further processing. The main calculations are: back testing (sample and out-of-sample), forecast, and advanced forecast. The output results are presented in a chart form and data table. Limitations:
· The free-trial version has some limited capabilities (files downloading and complete processing are disabled). The part of results can be explored using pre-calculated exemplary outputs.