Almost all data analysis software packages (SPSS, SAS, Statistical, Origin Pro, DataFit, Stata or Systat) need end-users to provide/guess initial start values, and the successes of nonlinear regression computation are havily depended on those guessed data. Unfortunately, for most of people, the guesses of proper start values are nightmare.

With the unique, robust and high efficient optimation algorithms of "Global Levenberg-Marquardt" and "Global BFGS", proposed newly, Auto2Fit is no long to require the end-users to provide/guess initial start values (but with free random ones), the performance, however, is much better than any others. Auto2Fit has turned to be the best one for nonlinear regerssion analysis nowadays.

Auto2Fit is also a flexible, versatile, easiest-to-use tool in the areas of function optimization, equation solving, parameter estimation, combinational problems (like TSP) and chart drawing.

Application Areas:

• Auto2Fit can be used in the areas such as:
  • Regression and Curve Fit: linear, nonlinear, multi-nonlinear with no limitations on variable and parameter number.
  • Global Optimization: find the maximal/minimal value of any function. The function may be nonlinear, non-continues, non derivable.
  • Equation Solve: solve any type of equation or system equations, both linear and nonlinear.
  • Model Auto-calibration: parameter estimation of complicated engineering model.
  • Combinational Problems: solve problems like TSP and Short-Path.
  • Chart Drawing: 2-D, 3-D cartesian and 2-D parametric function chart, especial for implicit functions.

• A number of optimization algorithms have been integrated in Auto2Fit:
  • Genetic Algorithm (GA): both encode/decode mode and real-value based mode, six crossover types and seven selection types.
  • Particle Swarm Optimization (PSO): six schemes (2 standard and 4 new).
  • Differential Evolution (DE): ten schemes (5 new) and four routs (3 new).
  • Maximum Inherit Optimization (MIO): four schemes.
  • Simulated Annealing (SA).
  • Simplex Method (SM).
  • Levenberg-Marquardt (LM): standard LM and Global LM.
  • Quasi-Newton (BFGS): standard BFGS and Global BFGS.
  • Self-Organizing Migrating Algorithms (SOMA).
  • Conjugate-Gradient Method (CGM).
  • Powell Optimization (OP).
  • Tabu Search (TS).

• Support VB and Pascal style script, as a result, any complicated problem could be described. The script engine is extremely fast, for example, 5 - 30 times faster than MS VB.
• Multi-data file, multi-outputs supported: this unique feature makes it very useful for parameter estimation with multi-data file.
• Easy to handle file-depended parameters and file-depended constants.
• Bulit-in Excel-like, multi-windows spreadsheet, easy for pre- and post-data process.
• Over 140 examples are included in the package, covering the areas of regression, curve fit, optimization, equation solving and parameter estimations.
• and much more...

Features include Binomial, Poisson, Normal, Exponential, Student t, Chi Squared, F, Beta and Lognormal Distributions; Inverses of Normal, Student t, Chi Squared, F, Beta and Lognormal Distributions; Lists of Binomial Coefficients, Factorials and Permutations; Calculations of Gamma and Beta Functions; Printing of Standard Normal Tables & Critical t Values; Fully Customisable; Integrated Help System which includes a Tutorial. We also plan on adding many more Distributions and features.

Also includes Graphing of Distributions, addition Info on Distributions and registered users also get Electronic Documentation and a PDF version of the docs designed for printing.

Ideal for the Maths/Stats Student who wishes to understand Probability Distributions better, as well as the Maths Buff who wants a well designed calculating tool.

Designed for Win32 and optimised for Windows XP and Vista.
 
Evaluation (time-limit)
Source: None
Price: $29

• Single, Dual (paired and unpaired) and Multiple Data Analysis (Multivariate analysis not in Lite Version).
• Data can be Time Based for Time Series Analysis.
• Data can be entered as Raw Data, Grouped Data, or as Summary Data.
• Raw Data can be converted into Grouped Data.
• Raw Data can have in-built Transformations applied to them.
• Raw Data can have Random values meeting user defined criteria.
• Data can be fully documented and stored in DataLists.
• Standardisation of Data around given mean/standard deviation.
• DataLists are grouped together in Workbooks - one Workbook is opened at a time and can contain many DataLists.
• Sample Size calculations.
• Random Value Lists for Sampling.
• Statistical Summary including: mean, median, mode, variance, standard deviation, kurtosis, skew, etc.
• Comparison of Statistics for Raw and Grouped of the same Data.
• Graphs and Charts including: Histograms, Line Graphs, Pie Graphs, Ogives, Scatter Diagrams, Rootograms, Bar Charts and more.
• Inference and Hypothesis Analysis of a Single Population - including the Mean, the Variance and Proportions.
• Inference and Hypothesis Analysis of Two Populations - including Difference of the Mean, Ratio of the Variance, Difference of Proportions.
• Analysis of Variance (ANOVA) (not in Lite Version).
• Linear Regression with Transformations of the dependent variable.
• Multiple Regression with Transformations of the dependent variables (not in Lite Version).
• Moving Averages.
• Trend Analysis of Time Series Data.
• PLUS MORE.

Features:

• Any more layer numbers, any more node numbers of each layer and any transfer functions of each layer.
• Visual design of network structures.
• Visual, real time and dynamic control of learning process.
• Fast converge of learning. For example, for "XOR" problem, learning error will converge to zero within three iterations.
• Steps-by-steps learning run mode.
• Easy settlement for time series analysis.
• Optimal analysis of inter-relationships and 2D-3D graphical evaluations of all factors.
• Support multi-learning data files. Files may be "Normal" and "Grid" types.
• VB, Java and Delphi/Pascal script languages supported.
• Excel-like, multi-windows data file editor, which may be also used as a stand-along spreadsheet.

• Discrete-time and continuous-time waveform synthesis using a variety of methods
• Frequency-domain transforms (DFT, FFT, DTFT, forward & inverse transforms, window functions)
• Convolution and correlation functions
• Solution of difference equations
• Analysis and design of analog filters, IIR and FIR digital filters
• Interactive pole-zero analysis of filters
• Digital filter simulation and code generation
• Power spectrum estimation using classical and modern techniques
• Powerful visual simulation capabilities for block-diagram based systems
• User-defined extensions using JScript and VBScript languages