How to find y-axis data for a certain x-axis value

I've tried to find IM_Collapse_find_S1M for 5km, 10 km, 15 km but it's giving wrong values. Actually values will increase gradually from 5km to 15 km but through this code I'm getting opposite nature. I don't know why I'm getting such errors.

load MIDR_S1M.xlsx
load PGA_EQ.xlsx 
PGA_Scale = [0.1:0.1:2.4];
xi = linspace(0,0.1,24);
MIDR_S1M_Rupture_5km = [MIDR_S1M(1,:);MIDR_S1M(4,:);MIDR_S1M(7,:);MIDR_S1M(10,:)];
MIDR_S1M_Rupture_10km = [MIDR_S1M(2,:);MIDR_S1M(5,:);MIDR_S1M(8,:);MIDR_S1M(11,:)];
MIDR_S1M_Rupture_15km = [MIDR_S1M(3,:);MIDR_S1M(6,:);MIDR_S1M(9,:);MIDR_S1M(12,:)];
 for tt = 1:12
     yi = interp1(MIDR_S1M(tt,:),PGA_EQ(tt,1)*PGA_Scale,xi,'nearest','extrap') ; 
     h1 = plot(xi,yi,'linewidth',1.5);hold on;
 end
 Desire_Limit_Collapse = 0.02; % values to be calculated from the graph corresponding this Desire_Limit_Collapse
for ii= 1:12
    IM_Collapse_find_S1M(ii)= interp1(MIDR_S1M(ii,:),PGA_EQ(ii,1)*PGA_Scale,Desire_Limit_Collapse,'nearest','extrap');
end
IM_Collapse_find_S1M_Rupture_5km = [IM_Collapse_find_S1M(1,1);IM_Collapse_find_S1M(1,4);IM_Collapse_find_S1M(1,7);IM_Collapse_find_S1M(1,10)];
IM_Collapse_find_S1M_Rupture_10km = [IM_Collapse_find_S1M(1,2);IM_Collapse_find_S1M(1,5);IM_Collapse_find_S1M(1,8);IM_Collapse_find_S1M(1,11)];
IM_Collapse_find_S1M_Rupture_15km = [IM_Collapse_find_S1M(1,3);IM_Collapse_find_S1M(1,6);IM_Collapse_find_S1M(1,9);IM_Collapse_find_S1M(1,12)];

Answer 1

[dpb]

Your X variable is repeating in sets of three so when you use all 12 elements as a vector it's multi-valued in y. interp1 then picks the first that it finds for any given Xi

The overall curve then looks like the following--

Above is from

MIDR_SIM=readmatrix('MIDR_S1M.xlsx');
PGA_EQ=readmatrix('PGA_EQ.xlsx'); 
plot(MIDR_SIM(:,1),PGA_EQ)

for the first column. You'll need to interpolate over the individual segments within the overall sets of 12 to avoid the multi-valued Y values to get the individual pieces.

[Sumit Saha]

okay...but whenever I'm trying to find for 5km and 10 km ....still 5 km is higher....is there any issue with the interpolation function ?

[dpb]

Show your revised code that doesn't try to use the whole thing at once -- it's the root of the issue methinks...

[Sumit Saha]

IM_Collapse_find_S1M_Rupture_5km = [1.077 0.877 0.657 0.607];

IM_Collapse_find_S1M_Rupture_10km = [1.699 1.31 0.96 0.879];

IM_Collapse_find_S1M_Rupture_15km = [2.716 1.939 1.4 1.329];

see this thing I got from excel interpolation and the trend of this data is right with my model. But if I remove the for end loop in this code and try to calculate it from individual plot (MIDR(1,:),PGA_EQ(1,1)*PGA_Scale) then also it's showing the wrong trend for this data points.

    IM_Collapse_find_S1M = interp1(MIDR_S1M(1,:),PGA_EQ(1,1)*PGA_Scale,Desire_Limit_Collapse,'nearest','extrap');

[dpb]

I still don't follow what you're trying to do here -- can you explain the end objective?

What I do see is that for MIDR_S1M(1,:); there's a fairly smooth curve for first 22 points then the 23rd point is some value of 10E18! Whassup w/ that?