Mean value of timeseries data in matrix form

Dear all, I have very basic question, I have time series for one variable (change between -1 and +1) with the same length (about 10000 second) in a matrix form (A), I calculate the mean with this way : mean_A = mean (A), where A is a matrix (1000 X 100)but the results very strange for me.

Thanks in advance

[Birdman]

Can you share what you get from your code? Also please share your entire code.

[Student for ever]

M=mean(T5_8); where T5_8 is a matrix (1000 X 100)

T5_8=[eta1 eta2 eta3 eta4 eta5 eta6 eta7 eta8 eta9 eta10 eta11 eta12 eta13 eta14 eta15,... eta16 eta17 eta18 eta19 eta20 eta21 eta22 eta23 eta24 eta25 eta26 eta27 eta28 eta29 eta30,... eta31 eta32 eta33 eta34 eta35 eta36 eta37 eta38 eta39 eta40 eta41 eta42 eta43 eta44 eta45,... eta46 eta47 eta48 eta49 eta50 eta51 eta52 eta53 eta54 eta55 eta56 eta57 eta58 eta59 eta60,... eta61 eta62 eta63 eta64 eta65 eta66 eta67 eta68 eta69 eta70 eta71 eta72 eta73 eta74 eta75,... eta76 eta77 eta78 eta79 eta80 eta81 eta82 eta83 eta84 eta85 eta86 eta87 eta88 eta89 eta90,... eta91 eta92 eta93 eta94 eta95 eta96 eta97 eta98 eta99 eta100]

and each eta has a length (1000)

[Birdman]

What does each column or row contain? Your -1 and +1 variable or something else?

[Birdman]

Share the weird result that you obtain.

[Student for ever]

eta is a column vary from -1 to +1

[Student for ever]

eta_mean

[Student for ever]

table of the matrix

Answer 1

[Birdman]

Actually the figure you obtain is not weird. It is an outcome of floating point number not expressed correctly in binary form, therefore it does not show as zero. Please read the following link to learn more about this situation and to learn its possible solutions.

https://www.mathworks.com/matlabcentral/answers/57444-faq-why-is-0-3-0-2-0-1-not-equal-to-zero