https://communities.sas.com/t5/SAS-Forecasting-and-Econometrics/What-DOes-convergence-criteria-satisfied-meaning-in-Pro-Logistic/m-p/427305#M2955 Fri, 12 Jan 2018 18:02:02 GMT ankur_1989 2018-01-12T18:02:02Z https://communities.sas.com/t5/SAS-Forecasting-and-Econometrics/What-DOes-convergence-criteria-satisfied-meaning-in-Pro-Logistic/m-p/427305#M2955 Fri, 12 Jan 2018 18:02:02 GMT https://communities.sas.com/t5/SAS-Forecasting-and-Econometrics/What-DOes-convergence-criteria-satisfied-meaning-in-Pro-Logistic/m-p/427305#M2955 ankur_1989 2018-01-12T18:02:02Z https://communities.sas.com/t5/SAS-Forecasting-and-Econometrics/What-DOes-convergence-criteria-satisfied-meaning-in-Pro-Logistic/m-p/427310#M2956 <P>Unlike ordinary least-squares regression, which is a direct method, many regression procedures have to solve nonlinear optimization problems in order to find the parameters in the model that best fit the data. The procedure starts with an initial estimate&nbsp;of the parameters and then iteratively refines that estimate until "convergence," which means that the parameters are optimal and further iteration will not improve the parameter estimates.</P> <P>.&nbsp;</P> Fri, 12 Jan 2018 18:20:47 GMT https://communities.sas.com/t5/SAS-Forecasting-and-Econometrics/What-DOes-convergence-criteria-satisfied-meaning-in-Pro-Logistic/m-p/427310#M2956 Rick_SAS 2018-01-12T18:20:47Z https://communities.sas.com/t5/SAS-Forecasting-and-Econometrics/What-DOes-convergence-criteria-satisfied-meaning-in-Pro-Logistic/m-p/427311#M2957 <P>It means the solution found is possibly not the solution, but the algorithm couldn't continue further.</P> <P>&nbsp;</P> <P>It could also mean you have too many model terms in your model, and removing some may allow the algorithm to find the solution.</P> Fri, 12 Jan 2018 18:21:30 GMT https://communities.sas.com/t5/SAS-Forecasting-and-Econometrics/What-DOes-convergence-criteria-satisfied-meaning-in-Pro-Logistic/m-p/427311#M2957 PaigeMiller 2018-01-12T18:21:30Z