Timetable_Averaging values from each month

1 view (last 30 days)
Leulaye Maskal
Leulaye Maskal on 21 Oct 2021
Commented: Leulaye Maskal on 22 Oct 2021
Timetable Averaging values from each month(Jan-Dec) from all years between 2000-2020. So I would like to have 12 USDM_Index values; one for each month over the 21 years
  2 Comments
Leulaye Maskal
Leulaye Maskal on 21 Oct 2021
I have tried that but its gives me the mean for each month within each yearh so I end up wit 12 * 21 for each month

Sign in to comment.

Answers (2)

Kelly Kearney
Kelly Kearney on 21 Oct 2021
I don't believe you can use retime to build a climatology; for that, I usually use the splitapply function. Unfortunately, splitapply's syntax for tables and timetables is very clunky (there's no easy way to apply the same function to all variables), hence my casting to and from arrays in the following example:
% The original timetable
T = timetable(datetime(2000,1:24,1)', rand(24,1), rand(24,1), ...
'VariableNames', {'USDM_index', 'other'});
% Calculate climatological monthly average
[g, mn] = findgroups(month(T.Time));
xclim = splitapply(@(x) mean(x,1), table2array(T), g);
% Reformat to timetable
refyr = min(year(T.Time)); % ... or whatever you want
Tclim = array2timetable(xclim, 'RowTimes', datetime(refyr,mn,1), ...
'VariableNames', T.Properties.VariableNames);

Duncan Po
Duncan Po on 22 Oct 2021
Use groupsummary with 'monthofyear' binning
T = timetable(datetime(2000,1:(12*21),1)', rand(12*21,1), 'VariableNames', {'USDM_index'});
S = groupsummary(T, 'Time', 'monthofyear', 'mean')
S = 12×3 table
monthofyear_Time GroupCount mean_USDM_index ________________ __________ _______________ 1 21 0.44367 2 21 0.44095 3 21 0.52971 4 21 0.66029 5 21 0.45509 6 21 0.59956 7 21 0.56916 8 21 0.49774 9 21 0.49873 10 21 0.49883 11 21 0.48883 12 21 0.56189

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!