How can I ort elements of a column in a matrix using elements of another column in another matrix

I have a matrix of 3 columns: A, B and C. The values of A and B are collected from floats. Each float collects data over a specific depth range by sinking several times. Thus, each sink produces a seperate depth profile. These profiles are assigned with numbers which are given in column C (if each float generates 10 profiles, then for the first float, C = 1 to 10; for the second float, C = 11 - 20 and so on...). For every C, there is a set of values for A and B. Hence my data look like:

Matrix 1

A B C

24 35 1

17 35 1

11 35 1

5 35 2

26 34 2

20 34 2

15 34 3 so on...

A and B correlate with one another. Therefore I wanted to find y-intercepts of the A-B regression line for each profile. If we take the above dataset, for the three profiles shown, there will be three y-intercepts. Thus, my results look like:

Matrix 2

Profile Y-intercept

1 xxx

2 - xxx

3 xxx

IMPORTANT: The length of Matrix 2 is shorter than length of Matrix 1

The calculated y-intercepts can be either positive or negative. I am only interested in the positive values. I have already isolated the positive y-intercepts and the profiles that are responsible in generating them using an if loop. The next step I want to achieve is to go to the Matrix 1 and only keep rows where the value of C equals to the selected profile numbers. So I want to see something like this:

Matrix 3

A B C

24 35 1

17 35 1

11 35 1

15 34 3 (because profile number 2 gives a negative y-intercept, I want all the rows corresponding to C = 2 eliminated).

I can manually create Matrix 3 on an Excel by hand picking the rows based on the profiles. But I have data from 6098 profiles with a total of 29943 rows. Therefore I would very much like to know if there is any easier way of doing this on Matlab.

Thank you in advance.

Answer 1

[Matt J]

subset=[1,3] %The vector of IDs with positive y-intercept

idx=ismember(Matrix1(:,3),subset);

Matrix1=Matrix1(idx,:);

[Amavi Silva]

It worked! Extremely thank you!