I have the Y co-ordinate to a very complex graph. How do I extract and plot the corresponding X coordinates?

I can't generate the function of the graph because its too complex. I've plotted the graph using input data points. I have particular random values on the Y-axis and I need the corresponding X values. the find() function doesn't work because I think it only returns integral indices (Mine won't be an integral index).

[Image Analyst]

How did you plot it with only the Y data? If you don't have any x values, then of course it will just use the index as the x values. If you don't have the formula, then there is no way you can get any x other than the index. I mean, how could you possibly know what they are? Let's say I gave you Y = [34,56,83,155]. It will assume x = [1,2,3,4]. How could it possibly know that the x I was thinking of was [2,4,6,8]? Or maybe I was thinking x was really [4,13,25,68]. There's just no way of telling the x values at all, if all you have are the Y values and no equation or actual x data.

[Samar Kenkre]

I do have X values that's how i plotted it, but I need a specific X value for a given Y value which isn't a part of my input data

Answer 1

[quicklearner]

If you have the plotted graph, then perhaps you can use the grabit.m filw ehich is available in the Matlab file exchange for free. It helps you extract data points from a graph.

All the best !

Answer 2

[Star Strider]

‘I do have X values that's how i plotted it, but I need a specific X value for a given Y value which isn't a part of my input data’

Assuming your data meets the requirements of the interpolation functions, I would use the interp1 function. To get an x for a specific y, this will likely work:

x_new = interp1(y, x, y_new)

with x and y the data you have, and y_new the data you want the corresponding x value (here x_new) for.

[Samar Kenkre]

It says the function has to be monotonically increasing, which in my case isn't. But the function is pretty helpful! Thanks

[Star Strider]

One way is to force y to be monotonically increasing, then do the interpolation.

This works:

x = rand(1,10);                     % Create Data
y = rand(1,10);                     % Create Data
% 
y_new = 0.5;                        % Known ‘y’ Value
[ys,yi] = sort(y);                  % Force ‘y’ To Be Monotonically Increasing
x_new = interp1(ys, x(yi), y_new);
% 
figure(1)
plot(x, y, '+r')
hold on
plot(x_new, y_new, 'bp')
hold off
grid

[Walter Roberson]

If your y is not monotonic then you have the risk that asking to look up a particular y value would return multiple results.