convert hourly data to daily data

Hi, I have a hourly time series and I want to convert them to daily data so that the daily data be the avarage of hourly data.

Also, I have some gap data. I want when there is no data in a specific hour, the code does not consider that especific time and just avarage the rest of that 24 hours to convert to avarage daily data.

[Armin Azad]

I would be more than gratful if you could help me for that

Answer 1

[Star Strider]

Try this —

T1 = readtable('https://www.mathworks.com/matlabcentral/answers/uploaded_files/829100/Norwalk2_012095493_ready.xlsx', 'VariableNamingRule','preserve')

T1 = 104445×2 table
          datetime          discharge
    ____________________    _________

    01-Jan-2010 00:30:00      11.4   
    01-Jan-2010 01:30:00      11.4   
    01-Jan-2010 02:30:00      11.1   
    01-Jan-2010 03:30:00      9.74   
    01-Jan-2010 04:30:00      10.1   
    01-Jan-2010 05:30:00      10.1   
    01-Jan-2010 06:30:00      9.74   
    01-Jan-2010 07:30:00      10.1   
    01-Jan-2010 08:30:00      10.4   
    01-Jan-2010 09:30:00      10.4   
    01-Jan-2010 10:30:00      9.12   
    01-Jan-2010 11:30:00      9.74   
    01-Jan-2010 12:30:00      9.43   
    01-Jan-2010 13:30:00      10.1   
    01-Jan-2010 14:30:00      9.74   
    01-Jan-2010 15:30:00      9.43   

TT1 = table2timetable(T1)

TT1 = 104445×1 timetable
          datetime          discharge
    ____________________    _________

    01-Jan-2010 00:30:00      11.4   
    01-Jan-2010 01:30:00      11.4   
    01-Jan-2010 02:30:00      11.1   
    01-Jan-2010 03:30:00      9.74   
    01-Jan-2010 04:30:00      10.1   
    01-Jan-2010 05:30:00      10.1   
    01-Jan-2010 06:30:00      9.74   
    01-Jan-2010 07:30:00      10.1   
    01-Jan-2010 08:30:00      10.4   
    01-Jan-2010 09:30:00      10.4   
    01-Jan-2010 10:30:00      9.12   
    01-Jan-2010 11:30:00      9.74   
    01-Jan-2010 12:30:00      9.43   
    01-Jan-2010 13:30:00      10.1   
    01-Jan-2010 14:30:00      9.74   
    01-Jan-2010 15:30:00      9.43   

TT1dm = retime(TT1,'daily',@(x)mean(x,'omitnan'))

TT1dm = 4352×1 timetable
     datetime      discharge
    ___________    _________

    01-Jan-2010     9.8246  
    02-Jan-2010        NaN  
    03-Jan-2010        NaN  
    04-Jan-2010     7.5137  
    05-Jan-2010     7.1225  
    06-Jan-2010     6.7246  
    07-Jan-2010     6.4671  
    08-Jan-2010     6.2942  
    09-Jan-2010     5.9283  
    10-Jan-2010        NaN  
    11-Jan-2010        NaN  
    12-Jan-2010     4.5675  
    13-Jan-2010     4.3863  
    14-Jan-2010     4.1396  
    15-Jan-2010     4.1513  
    16-Jan-2010     4.4896  

TT1ds = retime(TT1,'daily',@(x)std(x,'omitnan'));
N = size(TT1dm,1);
dt = TT1dm.datetime;
dts = string(dt);
mv = TT1dm.discharge;
sv = TT1ds.discharge;
xv = 1:numel(dts);

figure
semilogy(xv, mv, '-b')
hold on
plot(xv, mv+(sv/sqrt(N))*1.96, ':r')                            % Optional
plot(xv, mv-(sv/sqrt(N))*1.96 , ':r')                           % Optional
hold off
grid
xt = xticks;
xtv = fix(linspace(min(xv), max(xv), numel(xt)));
set(gca, 'XTick',xtv,  'XTickLabel',dts(xtv))
xlim([min(xtv) max(xtv)])
legend('Daily Mean','±1.96 SEM', 'Location','best')

.