Chaotic Time Series Prediction using RBF Neural Networks

MATLAB 10K 206 views 0 downloads 1 credits
Tags:

Resource Overview

RBF-based prediction of chaotic time series implemented in MATLAB, using one-dimensional data generated from Lorenz time series with neural network architecture implementation details

Detailed Documentation

Chaotic time series prediction using RBF neural networks is a methodology that employs Radial Basis Function neural network models for time series forecasting. This implementation utilizes MATLAB programming environment, where we generate experimental data from the Lorenz chaotic system to create one-dimensional time series data. The implementation typically involves configuring the RBF network architecture with appropriate hidden layer neurons, selecting suitable radial basis functions (such as Gaussian functions), and implementing training algorithms like least squares method for weight optimization. Through this approach, we can effectively analyze and predict complex behaviors and trends in chaotic time series, with MATLAB providing built-in functions for network creation (newrb), training (train), and prediction (sim) operations.

You May Also Like