Select the analysis options for Nonlinear Regression

Stat > Regression > Nonlinear Regression > Options

Weights

Enter a numeric column of weights to perform weighted regression. Weighted regression is a method that can be used when the least squares assumption of constant variance in the residuals is violated (also called heteroskedasticity). With the correct weight, this procedure minimizes the sum of weighted squared residuals to produce residuals with a constant variance (also called homoskedasticity). For more information about determining the appropriate weight, go to Weighted regression.

The weights must be greater than or equal to zero. The weights column must have the same number of rows as the response column.

Confidence level for all intervals

Enter the level of confidence for the intervals that are likely to contain the parameter estimates and the predictions. Usually, a confidence level of 95% works well. A 95% confidence level indicates that if you took 100 random samples from the population, the confidence intervals for approximately 95 of the samples would contain the estimated value. For a given set of data, a lower confidence level produces a narrower interval, and a higher confidence level produces a wider interval.

Algorithm

To estimate the parameters, Minitab uses an algorithm to converge on the minimum sum of the squared residuals (SSE). For more information, go to Understanding algorithms and starting values in nonlinear regression.

Gauss-Newton: Select the Gauss-Newton algorithm.
Levenberg-Marquardt: Select the Levenberg-Marquardt algorithm.

Maximum number of iterations

Enter the maximum number of iterations that the algorithm can use to achieve convergence. Usually, the default value works well. If the algorithm fails to converge, you can increase the maximum number. However, you might need to change the algorithm, expectation function, or starting values to obtain a solution.

Convergence tolerance

Enter the convergence tolerance. Usually, the default value works well.