Calculating cross-point within a set of x & y data

I have a set of x & y data that I collected from a test. The plot of the data resembles a mountain with a single peak. Given this set of data, I want to calculate the values for x where y is 98% of the peak (max) value. There are two values for x. Any suggestions.

x =

   -0.0004
    0.0521
    0.1047
    0.1573
    0.2100
    0.2630
    0.3152
    0.3687
    0.4200
    0.4753
    0.5261
    0.5791
    0.6313
    0.6830
    0.7365
    0.7896
    0.8431
    0.8943
    0.9474
    1.0004

y =

   17.9863
   18.7457
   19.6454
   20.7195
   21.8660
   23.0983
   24.7895
   26.6830
   29.0743
   31.8048
   35.1163
   39.2968
   44.3394
   51.4311
   58.5378
   51.3318
   42.9399
   37.2848
   33.2080
   30.2815

Answer 1

[Sean de Wolski]

[mxy idx] = max(y); %max and index
y98 = 0.98*mxy;  %value of y crossing

pt1 = interp1(y(1:idx),x(1:idx),y98);  %interp lower value
pt2 = interp1(y(idx:end),x(idx:end),y98); %interp higher value

plot(x,y,'b-',[pt1 pt2],[y98 y98],'gd') %visualize

Edit for multiple x-ings Per comments:

x = cumsum(repmat(x,3,1));  %sample data your x/y from before
y = repmat(y,3,1);

y98 = 0.98*max(y);
yxing = y-y98;  %zero the yvalue
ylox  = find(diff(sign(yxing))); %crossings
ylox(:,2) = ylox+1; %both sides
ylox = ylox(:);
xvals = reshape(x(ylox),[],2); %set up poits
yvals = reshape(y(ylox),[],2);
n = size(xvals,1);
xings = zeros(n,1);

for ii = 1:n
  xings(ii) = interp1(yvals(ii,:),xvals(ii,:),y98);  %interp each point
end    

plot(x,y,'b-',xings,y98,'gd') %visualzie

[Sonoma Rich]

This will work for this data set, but what if the data set had two or three humps (peaks) with multiple cross-points? Any suggestions?

Answer 2

[Oleg Komarov]

http://www.mathworks.com/matlabcentral/fileexchange/11837-fast-and-robust-curve-intersections

http://www.mathworks.com/matlabcentral/fileexchange/22441-curve-intersections

And many other on the FEX if you look for intersection