If your equation uses one of these new rules, Prism 4 might not be able to find a reasonable fit until you tweak those initial values. Prism 5 offers more rules for defining initial parameter values. If one of the models is ambiguous, then Prism chooses the other model, without doing the F test or AIC comparison. When you compare two models, Prism 5 does an extra step. Since the weighting is computed differently, you can't directly compare the weighted sum-of-square values reported by the two versions of Prism. ![]() The method used by Prism 5 is better, so the results of Prism 5 are more correct. ![]() Prism 5 weights by the Y value of the curve, while Prism 4 (and earlier releases) weighted by the Y value of the data. If you chose to weight by the Y values (or the Y values squared), Prism 5 handles weighting differently than did Prism 4. The differences, if any, are usually trivial. Prism 5 has a few improvements in the fitting algorithm, so occasionally it can find a better fit than did Prism 4. The goal of regression is to minimize that sum of squares, so see which version of Prism found a fit with the smaller sum-of-squares. If you chose no weighting, check the sum-of-squares from the two programs. Prism 4 presented a full set of results in this case, but the results are not useful when the fit is ambiguous. If your fit is labeled "Ambiguous" by Prism 5, you know that some of the parameters are not determined precisely. Prism 5 does not use precisely the same algorithm as did Prism 4, so curve fitting results can be different in rare cases:
0 Comments
Leave a Reply. |
Details
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |