Prev: Right approach for using ROC curve (perfcurve) ?
Next: Face recognition technique for newbie
From: Luigi Giaccari on 7 Feb 2010 04:58
"Eric Guild" <eric.guild(a)afit.edu> wrote in message <h161em$en7$1(a)fred.mathworks.com>...
> Hi all,
> I have two groups of data with a mean and a variance, which I named M1/M2 and V1/V2. I plotted the Gaussian distribution of these groups of data by using this formula:
>
> G1 = (1/sqrt(2*pi)*V1)*(exp(-(x - M1).^2./(2*V1.^2)));
> G2 = (1/sqrt(2*pi)*V2)*(exp(-(x - M2).^2./(2*V2.^2)));
>
> Now, I want to get the Receiver Operating Characteristic curves for the two Gaussians. Namely, I want to move from left to right along the x-axis taking the area at each 'x' point of both the Gaussians, and plot them against eachother, so it will form a ROC curve from zero to one, since one is the maximum area under the Gaussian. Please Help me to figure out how to plot the ROC curve.
>
> Thanks,
> Eric

http://www.advancedmcode.org/roc-curve.html
 | 
Pages: 1
Prev: Right approach for using ROC curve (perfcurve) ?
Next: Face recognition technique for newbie