MATLAB: How to sum up numbers in time series from .txt file

time series

I have a .txt file with time series matrix like this: 
2014,01,29,14,04,0.200
2014,01,29,14,50,0.000
2014,01,29,15,50,0.000
2014,01,29,16,50,0.000
2014,01,29,17,50,0.000
2014,01,29,18,50,0.000
2014,01,29,19,50,0.000
2014,01,29,20,50,0.000
2014,01,29,21,50,0.000
2014,01,29,22,50,0.000
2014,01,29,23,50,0.000
2014,01,30,0,50,0.000
2014,01,30,01,15,0.787
2014,01,30,01,50,0.000
2014,01,30,01,51,0.200
2014,01,30,02,49,0.589
2014,01,30,02,50,0.200
2014,01,30,03,03,0.393
2014,01,30,03,22,7.088
2014,01,30,03,33,6.101
2014,01,30,03,44,10.358
2014,01,30,03,50,2.952
2014,01,30,03,55,1.969
2014,01,30,04,02,0.982
2014,01,30,04,09,0.200
The columns show year, month, day, hour, minute and precipitation rate, respectively. I need hourly precipitation data. I want to get output like this in a .txt file:
2014,01,29,14,0.200
2014,01,29,15,0.000
2014,01,29,16,0.000
...
And I need daily precipitation data in a other .txt file:
2014,01,29,0.200
2014,01,30,31.819
...
Tks for help!!

Best Answer

Load it using readtable. And do a splitapply or rowfun.
tbl = readtable('test.txt', 'ReadVariableNames', 0);
tbl.Properties.VariableNames ...
    = {'year', 'month', 'day', 'hour', 'minutes', 'precipationRate'};
[grp, hrTbl] = findgroups(tbl(:, {'year', 'month', 'day', 'hour'}));
hrTbl.preci = splitapply(@mean, tbl.(6), grp);
hrTbl.year  = splitapply(@(x) x(1), tbl.(1), grp);
And writetable it.
writetable(hrTbl, writePath);
Similar for your month and yearly table.
Related Question