Mean value every 60 rows

Hello, I got my table, with 180 rows and 2 columns, each column is a variable X1, X2. I want to have the mean of X1 60 rows in 60 rows, the same for X2. This is the result media =

   60.1635
   60.1523
   60.1492

media =

513.0252
513.1533
513.1119

Question 1) How to store the results in a matrix or table 3x2 so I would have something like:

   60.1635 513.0252
   60.1523 513.1533
   60.1492 513.1119

Question 2) For example if i wanted to make the average each 120 rows, what would I need to change? is just the "n" value? because i dont understand what does the "1" do in here

      a = reshape(tabela(:,j),[],1);

Code is here:

n = 60

for j=1:2
    a = reshape(tabela(:,j),[],1);
    media = arrayfun(@(i) mean(tabela(i:i+n-1,j)),1:n:length(a)-n+1)'
end

Thanks for your help!

Answer 1

[Birdman]

You need to do some tricks here. Firstly, your data is 181x2 and since 181 is not dividable by 60, I erased the last row to make it 180x2. Next do the following steps:

Tabela=mat2cell(tabela,60*ones(1,3),2)
Tabela_avg=cellfun(@(x) mean(x),Tabela,'uni',0)

and you will obtain the mean for each 60x2 data. You can reach all of them at once by typing

Tabela_avg{:}

If you want the first 120 rows, then you need to separate that as follows:

Tabela=mat2cell(tabela,[120 60],2)

since the sum of dimensions has to be equal to the initial size.

[Tiago Dias]

Oh yeah sorry, 181 my bad.

Could you explain to me what each line does, so i can see the difference from mine.

I asked my Question2 wrongly sorry, what to change in my code or in yours, in the case for me to have each 10 rows a mean, so in that case i would end up with 18 values of mean for each variable

[Birdman]

Tabela=mat2cell(tabela,60*ones(1,3),2)

Line above separates your initial matrix into 3 matrices having sizes of 60x2 and they are stored in cells.

Tabela_avg=cellfun(@(x) mean(x),Tabela,'uni',0)

This line automatically calculates the mean values for your 3 60x2 sized matrices. You can see them by typing:

Tabela_avg{1}
Tabela_avg{2}
Tabela_avg{3}

[Tiago Dias]

I express myself wrong, i want a double 3x2 display, so if i want the 1st mean of X1 i call Tabela_avg(1,1) and it comes = 60.1635

[Birdman]

Then type,

Tabela_avg=cell2mat(cellfun(@(x) mean(x),Tabela,'uni',0))

[Tiago Dias]

oh yeah dumb me, 3 is the number os blocks created, thats good because i can calculated them previously so instead of 3 i can put a letter or something. for example if it was a mean for each 10 rows would it be:

Tabela=mat2cell(tabela,10*ones(1,18),2)

can it be in doubles? so each mean goes to i =1 , 2 and 3 for j=1...and for j=2 the same?

[Birdman]

This way it will be double matrices stored in a cell. You can extract them by folloing notation:

Tabela{1}
.
.

[Tiago Dias]

This does what i want, so thanks for your time and patience!!

    A = cell2mat(Tabela_avg)

[Tiago Dias]

Sorry to bother again, but what does the 'uni' and '0' do?

[Stephen23]

"what does the 'uni' and '0' do?"

These are short for 'UniformOutput' and false.

They tell cellfun to put the function outputs into a cell array and return this, which is useful in situations where the function outputs might not be uniform (i.e. are non-scalar). The default ('UniformOutput' = true) tells cellfun that the function outputs are scalar (so they can be joined together into an array the same size as the input cell arrays). See the cellfun help for more info.

Answer 2

[KL]

Simpler:

No need to use loops/arrayfun or cellfun at all. Simpler solution is to use just reshape with mean,

 cc = reshape(mean(reshape(dummy,n,[])),[],2)
 cc =
   60.1635  513.0252
   60.1523  513.1533
   60.1492  513.1119

[Tiago Dias]

I express myself wrong, i want a double 3x2 display, so if i want the 1st mean of X1 i call Tabela_avg(1,1) and it comes = 60.1635

[KL]

The answer shown above is 3x2 matrix of double!

[Tiago Dias]

in what u wrote I only see

   cc = reshape(mean(reshape(dummy,n,[])),[],2)

so what is dummy?

[KL]

That's an example Tiago! You should replace dummy with your matrix.

[Tiago Dias]

OK, thanks !

Answer 3

[sura Naji]

l have amatrix contact from 1 colum and 1000 rows l want to find the mean and the standard deviation also l want to find the worst value of this because l use the particle swarm optimization