NLR

NLR and CNLR are available in Custom Tables and Advanced Statistics.

Nonlinear regression is used to estimate parameter values and regression statistics for models that are not linear in their parameters. There are two procedures for estimating nonlinear equations. CNLR (constrained nonlinear regression), which uses a sequential quadratic programming algorithm, is applicable for both constrained and unconstrained problems. NLR (nonlinear regression), which uses a Levenberg-Marquardt algorithm, is applicable only for unconstrained problems.

MODEL PROGRAM parameter=value [parameter=value ...]
transformation commands

[DERIVATIVES
transformation commands]

[CLEAR MODEL PROGRAMS]

Procedure CNLR (Constrained Nonlinear Regression):

[CONSTRAINED FUNCTIONS
transformation commands]
 
CNLR dependent var
 
 [/FILE=file]   [/OUTFILE=file]
 
 [/PRED=varname]
 
 [/SAVE [PRED] [RESID[(varname)]] [DERIVATIVES] [LOSS]]
 
 [/CRITERIA=[ITER n] [MITER n] [CKDER {0.5**}] 
                                      {n    }
            [ISTEP {1E+20**}] [FPR n] [LFTOL n] 
                   {n      }
            [LSTOL n] [STEPLIMIT {2**}] [NFTOL n] 
                                 {n  }
            [FTOL n] [OPTOL n] [CRSHTOL {.01**}]] 
                                        {n    }
 
 [/BOUNDS=expression, expression, ...]
 
 [/LOSS=varname]
 
 [/BOOTSTRAP [=n]]

Procedure NLR (Nonlinear Regression):

NLR dependent var
 
 [/FILE=file]   [/OUTFILE=file]
 
 [/PRED=varname]
 
 [/SAVE [PRED] [RESID [(varname)] [DERIVATIVES]]
 
 [/CRITERIA=[ITER {100**}] [CKDER {0.5**}] 
                  {n    }         {n    }
            [SSCON {1E-8**}]  [PCON {1E-8**}]  [RCON {1E-8**}]] 
                   {n     }         {n     }         {n     }

**Default if the subcommand or keyword is omitted.

This command reads the active dataset and causes execution of any pending commands. See the topic Command Order for more information.

Syntax for the NLR and CNLR commands can be generated from the Nonlinear Regression dialog box.

Example

MODEL PROGRAM A=.6.
COMPUTE PRED=EXP(A*X).

NLR Y.