How do I find standard deviations within bimodal data?

5 次查看（过去 30 天）
Dan Crompton
Dan Crompton 2016-9-22
评论： Dan Crompton 2016-9-22
I have a bimodal data set that looks something like this:
From this, I have been using the function fitgmdist to find the mixing porportions and means of the two peaks:
fit2 = fitgmdist(XM500,2)
fit2 =
Gaussian mixture distribution with 2 components in 1 dimensions
Component 1:
Mixing proportion: 0.418376
Mean: 0.4316
Component 2:
Mixing proportion: 0.581624
Mean: 0.2965
I can pull the amplitude of each peak from the data presented, but I would be unable to re-create the two curves without a standard deviation. Does anyone have experience in this and know how it may be possible?
Thanks!

请先登录，再回答此问题。

采纳的回答

Adam
Adam 2016-9-22
fit2.Sigma
will give you the covariance array which should just be variances if you have 1d data. The standard deviation can be calculated from this.

更多回答（0 个）

类别

在 Help Center 和 File Exchange 中查找有关 Gaussian Mixture Models 的更多信息

标签

产品

Community Treasure Hunt

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

Start Hunting!

Translated by