Fit (wrapped) gaussian distribution to circular data

Question

CP 2011-10-5

1
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/17532-fit-wrapped-gaussian-distribution-to-circular-data

编辑： Image Analyst 2020-11-13

I have a set of data that is distributed on a circle and I want to fit it to a normal distribution. What I mean is, the X axis of that plot is wrapped on a circle while the Y axis values are normally distributed. How would I go about doing this? A regular gaussian fit works fine for peaks in the middle of the circle, but when the peaks are near the seams of the circular dimension, fitting a regular gaussian distribution doesn't work very well.

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Daniel Shub 2011-10-6

2
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/17532-fit-wrapped-gaussian-distribution-to-circular-data#answer_23688

I think you might be looking for a von Mises function. I don't think MATLAB has any built in support for von Mises functions, but you should be able to code it.

2 个评论
显示无隐藏无

CP 2011-10-7

Slightly different properties than a wrapped gaussian but it's similar enough for my purposes and the function itself is simple enough to where doing a fit is much easier. Thanks!

Daniel Garside 2020-11-13

编辑：Daniel Garside 2020-11-13

There is a circular statistics toolbox that has Von Mises parameter estimation as part of it:

Code: https://github.com/circstat/circstat-matlab

Paper: doi.org/10.18637/jss.v031.i10

请先登录，再进行评论。

Answer 2

Image Analyst 2011-10-6

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/17532-fit-wrapped-gaussian-distribution-to-circular-data#answer_23689

编辑：Image Analyst 2020-11-13

Least Squares orthogonal distances fitting of circle, sphere, ellipse, hyperbola, and parabola.pdf

Try this citation:

"Least-squares orthogonal distances Fitting of circle, sphere, ellipse, hyperbola, and parabola" Sung Joon Ahn, Wolfgang Rauh, Hans-Jurgen Warnecke, Pattern Recognition 34 (2001) pages 2283-2303

Abstract: The least-squares fitting minimizes the squares sum of error-of-fit in predefined measures. By the geometric fitting, the error distances are defined with the orthogonal, or shortest, distances from the given points to the geometric feature to be fitted. For the geometric fitting of circle/sphere/ellipse/hyperbola/parabola, simple and robust nonparametric algorithms are proposed. These are based on the coordinate description of the corresponding point on the geometric feature for the given point, where the connecting line of the two points is the shortest path from the given point to the geometric feature to be fitted.

The paper is attached here.