How do I get decimal for the index of the function max()

Hi everyone,

I am using the fonction max() to reveal the maximum of 5 matrixes. I am using the code below:

[y1,x1] = max([max(WheelTorque_1) max(WheelTorque_2) max(WheelTorque_3) max(WheelTorque_4) max(WheelTorque_5)])
y1 =
   2.5798e+03
x1 =
     4

I checked the maximum on the matrixes and the y1 is correct but the x1 is false. The correct x1 is 4.85. I don't know if matlab does a truncation or if there is another problem.

Thanks for your help!

[KSSV]

Note that x1 is the index of y1 in the input array. It is not a value. So x1 says, your max value is from WheelTorque_4 which happens to be at 4th index.

[Monalisa Pal]

What exactly is required in x1? Same as Mr. KSSV said, the syntax which you followed is as follows:

[max_value_from_vector, index_of_max_value] = max(vector)

So your vector is [max(WheelTorque_1) max(WheelTorque_2) max(WheelTorque_3) max(WheelTorque_4) max(WheelTorque_5)], y1 is 2.5798e+03 and x1 is 4.

This means that y1 = vector(x1) = vector(4) = max(WheelTorque_4) = 2.5798e+03. Hope this helps you debug.

May be if you mention the functionality that you are trying to implement, we can help further.

[Max-Henri Froger]

Well thanks for your answers.

I am on app designer and I try to get the coordinate of the maximum point of the 5 curves (WheelTorque) and then plot the point on UIAxes. I am looking at the maximum point abscissa coordinate.

[KSSV]

So in that case, if you have absicca.....use

x(x1)

[Max-Henri Froger]

Yes but I don't know how to call to right matrix wheelTorque. Because the index say the maximum is in the WheelTorque_4 but I don't find the way to code this part. Do you know if I could use a system like WheelTorque_(i)(j) and i would be the index and j the x1? Because write like that it doesn't work

Answer 1

[John D'Errico]

The max function does NOT interpolate your data. Just because, IF you looked at the curve, you might decide the true maximum occurs someplace between between two points, max does not find that intermediate location.

Max returns to point in your array that is the maximum element, and the index of that element. No more. You cannot make it do more. Max is what it is. You cannot hope code will do what you want instead of what it is programmed to do.

If you truly need that intermediate point, then your main choice is to use a function that would smoothly interpolate the data. A spline is a common way to do that. And then use a tool like fminbnd or fminsearch to minimize the negative of your function. This will find a maximum.

If you have the underlying function that generated that values, you can use the same tools, fminbnd or fminsearch to minimize the negative of that function. Again, it will locate a max value then. But you cannot just use max.