how to find the data for same period of time ?

hi, I have two mat files, the first column of both files is date-time and the second column is data. I want to compare both the data sets. however, all dates and time are not same. here I want to find data from both files which has same measurement date and time. other data I won't need it. at last, i would get one mat file which would have date-time, data from first file and data from the second file. any idea how to make it possible ?

Answer 1

[Ameer Hamza]

Try this

data1 = load('MRRarranged.mat');
data2 = load('distro-data.mat');
dates1 = data1.MRR_arranged_data(:,1);
dates2 = data2.rawdata(:,1);
indexDates2 = ismember(dates2, dates1);
indexDates1 = ismember(dates1, dates2);
finalDates = data2.rawdata(indexDates2, 1);
finalData1 = data1.MRR_arranged_data(indexDates1, 2);
finalData2 = data2.rawdata(indexDates2, 2);
final = [table(finalDates) table(finalData1) table(finalData2)];

then you can save the variable final.

[pruth]

hi ameer, thanks for the reply again. I don't understand how this code will give me similar dates from both files, and corresponding data.

 finalDates = data2(index, 1);
finalData = data2(index, 2);

it will give me data only from data2, not from data1 right? shall I write this?

 finalData1 = data1(index, 2);

however the code stuck after-- finalDates = data2(index, 1);

it gives me an error saying Index exceeds matrix dimensions. what does it mean?

[Ameer Hamza]

I oversaw that you want data from both files, I have updated the answer. Since the dates are common. You can either read it from file1 or file2 it does not matter/

[pruth]

Ameer, that works as expected. thanks a lot. I find it interesting to play with Matlab commands. I found another command which also should work.

 [C,ia,ib] = intersect(rawdata,MRR_arranged_data,'rows')

i tried this but it gives me empty matrix. i dont know why ? anyway your code works perfect.

[Ameer Hamza]

The option 'rows' only compares row-wise

A   B
1   2  <- does not intersect
2   2  <- intersect
3   1  <- does not intersect
1   1  <- intersect

Since you don't want this behavior, use the command like this

   [C,ia,ib] = intersect(rawdata,MRR_arranged_data)

Answer 2

[Peter Perkins]

It may be that join on timetables is the way to go here.

>> load('distro-data.mat')
>> load('MRRarranged.mat')
>> tt1 = array2timetable(MRR_arranged_data(:,2:end),'RowTimes',datetime(MRR_arranged_data(:,1),'ConvertFrom','datenum'))
tt1 =
  15314×3 timetable
            Time            Var1    Var2    Var3
    ____________________    ____    ____    ____
    05-Apr-2018 00:00:02     0       0       0  
    05-Apr-2018 00:00:12     0       0       0  
    05-Apr-2018 00:00:22     0       0       0  
    05-Apr-2018 00:00:32     0       0       0  
    05-Apr-2018 00:00:42     0       0       0  
[snip]
>> tt2 = array2timetable(rawdata(:,2:end),'RowTimes',datetime(rawdata(:,1),'ConvertFrom','datenum'))
tt2 =
  2880×3 timetable
            Time            Var1    Var2     Var3 
    ____________________    ____    ____    ______
    05-Apr-2018 10:08:59     0       0      0.0049
    05-Apr-2018 10:09:59     0       0      0.0049
    05-Apr-2018 10:10:59     0       0      0.0049
    05-Apr-2018 10:12:00     0       0      0.0049
    05-Apr-2018 10:13:00     0       0      0.0049
[snip]
>> tt12 = innerjoin(tt1,tt2,'Key','Time')
tt12 =
  69×6 timetable
            Time            Var1_tt1    Var2_tt1    Var3_tt1    Var1_tt2    Var2_tt2    Var3_tt2
    ____________________    ________    ________    ________    ________    ________    ________
    05-Apr-2018 10:12:00         0         0           0              0        0         0.0049 
    05-Apr-2018 10:13:00         0         0           0              0        0         0.0049 
    05-Apr-2018 10:14:00         0         0           0         0.0006        0          0.005 
    05-Apr-2018 10:18:00         0         0           0              0        0          0.005 
    05-Apr-2018 10:19:00         0         0           0              0        0          0.005 
    [snip]

I can't tell if this is what you are looking fior; there are only 69 matches. It may be that you really want some form of synchronize.