how can i split mtrix to three ?

hi

i have matrix m=153*71 i need to extract from it 3 matrices with dinsion 51*71 ?

Answer 1

[Voss]

Here is one possible way, if A is your matrix:

A1 = A(1:51,:);
A2 = A(52:102,:);
A3 = A(103:153,:);

Answer 2

[Image Analyst]

Here is a similar question and answer:

https://www.mathworks.com/matlabcentral/answers/1629495-divide-array-into-3-parts#answer_875085

[RADWAN A F ZEYADI]

thank you ,,i have matrix m 153 71

i need to exrtact 51 71 from m

when i do i obtain the same picture of the first matrix for example

m1=m(1:51,:);

m2=m(52:103,:)

m3=m(103:153,:)

i have obtained m3 and m2 the same pictures with m1 infact in the m matrix they are totlay differentmage from m1 infact on the whole matrix they are different

[Image Analyst]

Sorry that's not true. Look:

m = randi(99, 153, 5)

m = 153×5

72 2 19 99 1 46 35 44 11 85 11 25 58 1 45 69 11 81 94 35 42 9 20 54 31 76 96 69 66 64 9 14 63 52 62 96 31 10 95 36 68 72 93 3 98 30 97 65 97 99

m1=m(1:51,:)

m1 = 51×5

72 2 19 99 1 46 35 44 11 85 11 25 58 1 45 69 11 81 94 35 42 9 20 54 31 76 96 69 66 64 9 14 63 52 62 96 31 10 95 36 68 72 93 3 98 30 97 65 97 99

m2=m(52:103,:)

m2 = 52×5

19 8 41 41 39 91 16 7 58 51 78 10 94 82 5 93 65 95 71 67 4 60 78 28 61 75 91 41 55 28 68 5 20 92 6 70 52 43 73 24 73 87 83 1 41 47 99 85 63 98

m3=m(103:153,:)

m3 = 51×5

75 19 89 71 42 64 78 10 92 38 44 48 93 10 91 94 32 26 37 67 90 68 92 47 93 38 28 16 3 54 5 40 2 4 92 50 2 76 82 30 68 62 20 15 93 26 53 43 36 30

You can see that obviously they are not the same. You must have done something else to overwrite all the matrices with m.

Answer 3

[RADWAN A F ZEYADI]

thank you just the last thing i have 2 vectors and i want to compute the Confidence intervals 95"% for both of them how could be done ?

both of them 51*1 i upload the picture to check the figure

[Image Analyst]

I think it's hard or impossible to get 95% confidence limits when you have only 2 data points for each time point. You'd need dozens to get some good statistics. Then you could plot envelope curves where you have a 95% confidence that the curve will be between the two envelope curves.

[RADWAN A F ZEYADI]

i have exracted them from 2 matrixes 51*71 for both

[Image Analyst]

So you have either 51 or 71 curves? If so, for each time point you can pass those values into fitdist() to get the distribution and CDF from which you can get the 2.5% and 97.5% values of the CDF which enclose 95% of the values.

If you can't figure it out, attach the matrix in a .mat file.