Relative Ranking of position in an array

Hi!

I have a large array with 10 rows of numerical data. I am trying to find an 'ordering' by size of all the elements. I turned the raw data into an indexed array through maxk:

 [M10,I10]= maxk(dgoutput,10,1)

And my I10 array has the row number for the elements in order from greatest to least (1-10). It looks like this:

1	2	1	1	1	2	1	1	2
2	1	3	6	3	1	3	3	1
4	3	4	3	2	3	4	2	3
3	4	2	7	4	6	2	5	4
5	5	5	5	6	7	5	4	5
8	8	6	4	5	4	6	6	7
9	7	7	2	7	5	7	9	6
7	6	9	9	8	9	10	8	8
6	9	8	8	9	8	8	7	9
10	10	10	10	10	10	9	10	10

And I'm trying to find a way to count the occurrences when one number occurs in a higher row than another. As an example a

Count (1>2) would return 6 since there are 6 columns where 1 is above 2.

and Count(2>1) would return 3, etc.

Any help would be appreciated!

Answer 1

[David Hill]

If a is your matrix above.

sum(find(a==1)-find(a==2)<0);

[Mick Stukes]

Thanks for this! My situation is a touch more complicated than I had originally explained, and I'm struggling to adapt this method to my needs.

I am now searching through my array for columns where two specific criteria to be met:

1: the number 3 is in the first row

AND

2: the number 1 occurs in a higher row than the number 2

Currenlty I can only find it for exact placings with the code:

 AI = sum(I10(1, :) == 3 & I10(2, :) == 1 & I10(3, :) == 2);

Which gives me only the total where the rows have 3, 1 and then 2.

Can you help?

[Mick Stukes]

I have tried using the find command in conjunction with the & criterion:

 AI = sum(I10(1, :) == 3 & find(I10==1)-find(I10==2)<0);

but I get an error:

Requested 100000x100000 (9.3GB) array exceeds maximum array size preference. Creation of arrays greater than this limit may take a
long time and cause MATLAB to become unresponsive.

Answer 2

[Bruno Luong]

I10=[1	2	1	1	1	2	1	1	2
2	1	3	6	3	1	3	3	1
4	3	4	3	2	3	4	2	3
3	4	2	7	4	6	2	5	4
5	5	5	5	6	7	5	4	5
8	8	6	4	5	4	6	6	7
9	7	7	2	7	5	7	9	6
7	6	9	9	8	9	10	8	8
6	9	8	8	9	8	8	7	9
10	10	10	10	10	10	9	10	10];

[r1,~]=find(I10==1);
[r2,~]=find(I10==2);
count1gt2 = sum(r1<r2)

count1gt2 = 6

count2lt1 = sum(r2<r1)

count2lt1 = 3