optimum conditions - fitlm

1 回表示 (過去 30 日間)
Bran
Bran 2016 年 12 月 7 日
回答済み: John D'Errico 2016 年 12 月 7 日
I have used fitlm to fit a second order model to a set of data: Model = fitlm(X,Y,'y~1 +x1*x2 + x2+x1^2') The plot produces a 3D curve which I have depicted. Which is the easiest way to find the values for x1 and x2 which optimise y?

回答 (1 件)

John D'Errico
John D'Errico 2016 年 12 月 7 日
No, you have not depicted any curve or surface. You may think you attached something, but you did not. Do you see anything attached?
Anyway, what do you mean by OPTIMIZE? To optimize something, you need to decide what you are optimizing. At least, to get any help, you need to tell us. Are you minimizing something? Maximizing?
If your model is second order, thus a quadratic polynomial, then 1st semester calculus is sufficient, if all you want to do is maximize or minimize y given the now known coefficients. However, I would point out that the model you show has NO maximum or minimum. You can make y as large as you wish, so +inf, or as small, so -inf. So you need to be clear, because what you have said so far has no answer.

カテゴリ

Help Center および File ExchangeAnalysis of Variance and Covariance についてさらに検索

タグ

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by