Performs multivariate polynomial regression using the Least Squares method. The program determines the coefficients of the polynomial, the generalized correlation coefficient and the standard error of estimate.

Approximates continuous functions in 1 to 4 independent variables by finite power and/or trigonometric series. The program fits the function exactly at the grid points.

Approximates tabulated functions in 1 to 4 independent variables by finite power and/or trigonometric series. The program fits the function exactly at the grid points.

Freeware for fast training, validation, and application of regression/approximation networks including the multilayer perceptron, functional link network, piecewise linear network, self organizing map and K-Means. C source for applying trained networks. Extensive help. User-supplied txt-format training data files, containing rows of numbers, can be of any size. Pruning for approximate structural risk minimization.

Source code for fast training of multilayer perceptron for approximation. Example training and validation data files included. Source code for applying the trained network. Resulting networks can be pruned by NuMap7.06. The training algorithm can be modified to be substantially better than BP. For winzip password, go to :http://www-ee.uta.edu/eeweb/ip/Software/Software.htm

The program is based on the powerful model-free method of time series analysis Caterpillar-SSA (Singular Spectrum Analysis). The result of the Caterpillar-SSA processing is identification, analysis and forecast of additive components of time series (trends, periodicities, noise). The program can be applied to multivariate analysis/forecasting and change-point detection.

Performs linear regression using the Least Squares method. Determines the regression coefficients, the generalized correlation coefficient and the standard error of estimate. Does multivariate regression. Displays 2D and 3D plots.

FindGraph is comprehensive, feature-rich graph editor, as well as a curve fitting and digitizing tool for engineers, scientists and graduate students. Convert plots back to X,Y data from any published report for later processing and manipulations. The program offers 12 generic fits, including Linear Regression, Logistic Functions, Fourier approximation, Neural Networks, B-splines curves, plus a library of over 300 industry-specific formulas.

The CatMV program is a realization of the "Caterpillar"-SSA method for analysis of time series, which may contain missing values. The implemented algorithms result in extraction of additive components of time series such as trends and periodic components, with simultaneous filling in the missing data (if any). The program is able to perform forecasting if to add missing values after the last point of the time series.