Home                       
 Fractional Calculus
 Regression Tools    
 Statistical Tools     
 MESearch                
 Screensavers          
 Downloads              
 Euskaraz                 
 Links                       
 Contact                    


 
RegressionTools: The Program
 
Home
Contents  |  LR  |  LnR  |  ExpR  |  PowR  |  PR  |  MLR  |  MPR  |  NLR  |  More...
Contact

The program RegressionTools allows performing regressions that are far beyond the capabilities of the Online Regression Tools:

  • It can perform regressions without limitations in the number of variables and with hundreds of thousands and even millions of rows (the exact number depends on the regression type, number of variables, degree...).
  • It is a console program (for Windows) with the command-line parameters specifying the regression to perform, which allows executing it from other programs (for example to perform many regressions automatically). A launcher (batch file) is provided for an easier specification of the parameters and launching.
  • It reads the input from a file, shows the output and writes it to a file.
  • Supported regression types: Linear Regression (LR), Polynomial Regression (PR), Multiple Linear Regression (MLR), Multiple Polynomial Regression (MPR).
  • It has many term selection options, so that specific terms can be selected (for example excluding/including terms in a MPR).
  • It supports weighted regressions with any regression type.
  • It supports equality constraints (fixed points) with any regression type.
  • It supports least-norm regressions (selecting the norm and performing a minimization of the norm of residuals) with any regression type.
  • It shows many goodness of fit measures: Residual Sum of Squares (RSS), Coefficient of Determination (R2), in linear regression also the Correlation Coefficient (r), and in least-norm regressions also the Norm of Residuals (NR) and the Norm-Based Coefficient of Determination (R2norm).
  • It is endowed with other facilities like column masks to easily mask/unmask specific columns of the input file, variable naming, showing/storing input and output statistics (including the calculated value, the error and the relative error at each point)...

The above features can be combined to perform amazing regressions that allow developing powerful mathematical models, despite its lack of a fancy interface, as shows the following screenshot:

You can harness the power of RegressionTools in two ways:

  • By buying the program.
  • By buying regression (or mathematical modelling) services.

Write to xuru@xuru.org for further information.