Regression and Correlation of Wind Direction (circular) Data

Kind all,

I'm currently facing a problem which I thought simpler in the beginning. Basically I have outputs of wind direction from a model and observed wind direction from my sensors (both are in degrees from 0 to 359). What I would like to do is the equivalent of linear regression and correlation coefficient computation that it's used when comparing linear data.

What I would accomplish with a simple usage of polyfit and polyval cannot be done on circular data (in case of wind direction 0 is actually closer to 359 than to 20) and the r2 values obtained with the regular expression 1-(sum of squares of the residues / total sum of squares) are completely off.

Is there any MATLAB function for working on circular-circular correlation and regression?

I am already aware of the Circular Statistics Toolbox, but while it gives some way of computing correlations there's nothing for regression.

Thanks!

[Runa S]

Hi Federico. Did you solve this problem? I am having the same issue now, and am a bit lost.

[Jim Goodall]

I am also stuck...

[John Trimper]

Bump!

[Adam Danz]

circ_corrcc() from the circular statistics toolbox computes the circular correlation coeficient between two circular data.