Hi!
From the description, it appears to me that you have two tables t1, t2, with the columns A,B and value, output respectively. You compare the values of t2.value with the values in t1.A, and whichever difference is least, the corresponding t1.B value is stored in t2.output. The tiebreaker cases would be the lesser of the two values, and even further, the value with the lesser index.
I have assumed some random values for these 3 columns to illustrate the working with an example. The problem can be addressed with a simple and straightforward code as follows –
% Define the input columns (My random values)
A = [5 7 10]';
B = [5000 7000 10000]';
value = [6 8 9]';
output = zeros(3,1);
% Create the tables
t1 = table(A,B);
t2 = table(value,output);
% Algorithm starts here
for i=1:numel(t2.value)
diffVec = abs(t2.value(i) - t1.A);
least = min(diffVec);
indices = find(diffVec == least);
val = min(A(indices));
out = find(t1.A == val);
t2.output(i) = t1.B(out(1)); % out(1) is to counter the tiebreaker case,
% when t1.A has recurring values but with different corresponding t1.B values
end
