How to store corresponding two data to new matrix by using forloop?

Now,i have a matrix, like follows: The 'value' is correspond to the 'time' on the left.

     time1  value1  time2  value2
    0.82456  0.47  0.78049  0.44
    0.84626  0.40  0.80080  0.39
    0.86796  0.30  0.82052  0.31

and i want to make a new matrix by using these data. For example, if the 'time' 0.82 or more and less than 0.83, i want to store the corresponding 'value' in the same low(like follows).If there is no corressponding value, it's available to keep the cell as zero or vacant.

      time value1 value2
      0.78         0.44
      0.79    
      0.80         0.39
      0.81  
      0.82  0.47   0.31
      0.83
      0.84  0.40

I guess it's better to use for-loops and if statement, but i have no idea. If you some idea to do this, please help me. Thanks.

[per isakson]

"If there is no corresponding value" Fine, but if it's more than one value (of value1 or of value2) in an interval, what to do then?

[Sayaka Kamata]

Finally, i want to calculate the average of value like follows.

this table is example, actually, i have more than 40 samples (value1, value2,......value40...)

      time value1 value2 ... average
      0.78         0.44  ...  0.44
      0.79               ... 
      0.80         0.39  ...  0.39
      0.81               ...
      0.82  0.47   0.31  ...  0.39
      0.83               ... 
      0.84  0.40         ...  0.40

Answer 1

[KSSV]

data = [ 0.82456  0.47  0.78049  0.44
    0.84626  0.40  0.80080  0.39
    0.86796  0.30  0.82052  0.31];

time = [data(:,1:2:4)] ;
val = [data(:,2:2:4)] ;

ti = [ 0.78
    0.79
    0.80
    0.81
    0.82
    0.83
    0.84 ] ;

%%fix time to two decimals
time = round(time.* 10^2) ./ 10^2 ;
iwant = NaN(size(ti,1),2) ;

[ia,ib] = ismember(time,ti) ;
for i = 1:size(time,2)
    iwant(ib(ib(:,i)~=0,i),i) = val(ia(:,i),i) ;

end

[per isakson]

"if the 'time' 0.82 or more and less than 0.83" asks for floor rather than round

[Sayaka Kamata]

Thanks a lot! it solved!

Answer 2

[Akira Agata]

I would recommend using timetable and retime function, like:

% Convert data to datatable
time1 = [0.82456; 0.84626; 0.86796];
value1 = [0.47; 0.4; 0.3];
time2 = [0.78049; 0.8008; 0.82052];
value2 = [0.44; 0.39; 0.31];

TT1 = timetable(minutes(time1),value1);
TT2 = timetable(minutes(time2),value2);

% Uniform sampling time
time = [minutes(0.78):minutes(0.01):minutes(0.87)]';

% Resample the original data by the uniform sampling time
TT1a = retime(TT1,time,'firstvalue');
TT2a = retime(TT2,time,'firstvalue');

% Output
TTall = [TT1a, TT2a];

The output is as follows:

       Time     value1    value2
    ______    ______    ______
    0.78 分     NaN      0.44  
    0.79 分     NaN       NaN  
     0.8 分     NaN      0.39  
    0.81 分     NaN       NaN  
    0.82 分    0.47      0.31  
    0.83 分     NaN       NaN  
    0.84 分     0.4       NaN  
    0.85 分     NaN       NaN  
    0.86 分     0.3       NaN  
    0.87 分     NaN       NaN

[per isakson]

timetable was "Introduced in R2016b"

[Sayaka Kamata]

Thanks for your answer, it was helpful and seems to be very simple, but my MATLAB is not the up-to-date type,,, i wasn't able to use the code 'timetable'.