Nonlinear Estimation

  • Quick Logit Regression
    Creates logit regression for categorical (binary) dependent variables and a list of continuous predictor variables. Use the Generalized Linear Models facilities to perform stepwise and best subset selection of predictors in ANCOVA-like designs, or to analyze large analysis problems with many predictors.
  • Quick Probit Regression
    Creates probit regression for categorical (binary) dependent variables and a list of continuous predictor variables. Use the Generalized Linear Models facilities to perform stepwise and best subset selection of predictors in ANCOVA-like designs, or to analyze large analysis problems with many predictors.
  • User Specified Regression, Least Squares
    Fit arbitrary regression models using least squares estimation; you can specify a regression equation using standard notation (e.g., Var3=a+log(b*Var4)). Logical operators are also supported.

 Least squares estimation is aimed at minimizing the sum of squared deviations of the observed values for the continuous dependent variable from those predicted by the model. When using the least-squares criterion, the very efficient Levenberg-Marquardt algorithm can be used to estimate the parameters for arbitrary linear and nonlinear regression problems. For large data sets, when using the least-squares criterion, this is the recommended method for fitting nonlinear models.

 Note: If no model (syntax) is specified, Statistica will fit a simple linear model.
  • User Specified Regression and Loss Function
    Fit arbitrary regression models using custom-defined loss functions; you can specify a regression equation using standard notation (e.g., Var3=a+log(b*Var4)). Logical operators are also supported.

Statistica will estimate the parameters of the regression equation by minimizing a custom loss function, of the form Loss=Function (e.g., Loss=W*Abs(Obs-Pred)). Use the Least squares regression options (and the very efficient Levenberg-Marquardt algorithm) to estimate the parameters for arbitrary linear and nonlinear regression problems for large data sets (using the least-squares criterion; this is the recommended method for fitting nonlinear models.

 Note: If no model (syntax) is specified, Statistica will fit a simple linear model, using the least squares criterion.