How to select specific values and corresponding cell positions from 31 x 12 matrix?

Hi,

I have 31 x 12 matrix (A) with daily temperature for one year. Values range between 10.2 and 39°C.

How to find a new metrix B (? x 12) with all values (in each column) less than 1.15 times minimum value (ie. < 10.2*1.15) and corresponding cell positions?

Thanks in advance.

[Parthu P]

Hi,

I have 31 x 12 matrix (A) with daily temperature for one year. Values range between 10.2 and 39°C.

How to find a new metrix B (? x 12) with all values (in each column) less than 1.15 times minimum value (ie. < 10.2*1.15) and corresponding cell positions?

Thanks in advance.

Answer 1

[Fabio Freschi]

Because the number of rows in B is not fixed, you can store the result of your check in a cell array

% create the matrix
A = (39-10.2)*rand(31,12)+10.2;

% tolerance
tol = 1.15;

% get minimum
minA = min(A(:));

% cell array with required entries
B = arrayfun(@(i)A(A(:,i) < tol*minA,i),1:size(A,2),'UniformOutput',false);

You can now access your data using B{i}, where i is the index of the month

You can also get the pointers to the desired data using find

[iDay,jMonth] = find(A < tol*minA);

[Parthu P]

Hi Fabio, many thanks for respondng to my query.

However, it doesn't fully answer my question.

Let's say, column 1 of matrix A has 31 rows and minimum value is 10.2 and maximum value is 39. When threshold/tolerance is 1.15 times the minimum value (10.2), B { i } should have more than one row(s). But when I use above functions with different tolarance values, B { i } only show single row (1 x 12) output. But I need multiple rows in all 12 columns. What other functions I should try?

[Parthu P]

Sorry. It's B. It should have more than one row(s). But when I use above functions with different tolarance values, B only show single row (1 x 12) output. But I need multiple rows in all 12 columns. What other functions I should try?

[Fabio Freschi]

B has only one row, but each cell contains all values oq your query:

>> B

B =

  1×12 cell array

  Columns 1 through 6

    {5×1 double}    {2×1 double}    {0×1 double}    {2×1 double}    {[10.4252]}    {0×1 double}

  Columns 7 through 12

    {2×1 double}    {[10.7914]}    {2×1 double}    {4×1 double}    {[11.4044]}    {2×1 double}

as you can see, for example, B{1} has 5 entries.

How do you expect to have B if the different montsh can have a different number of days with temperature below your threshold? In other words, what is the size of the expexted output?

Another thing you can do is to have a 31 x 12 B matrix, where all values above the threshold are set to zero or NaN. In this case you can use

B = A;
B(A >= tol*minA) = 0;