Table with date and x need to plot graph

I have a table with two columns first column is the date and x is a value. I need to sum all the x values up for one day and then plot the cumulative summed x for each date.

Does anyone have any suggestions?

Answer 1

[Star Strider]

One approach:

[F,S,R] = xlsread('Matlabhelp.csv');
Final = table(F(:,1),F(:,2),F(:,3), 'VariableNames',{'Date','X','Y'});
Final.Date = datetime(Final.Date,  'ConvertFrom','excel', 'Format','dd-MMM-yyyy')
[G,TID] = findgroups(Final.Date);
FinalSummed.Date = TID;
FinalSummed.X = splitapply(@sum, Final.X, G);
FinalSummed.Y = splitapply(@sum, Final.Y, G);
FinalSummed = struct2table(FinalSummed)

producing first:

Final =

    5×3 table

         Date          X        Y   
      ___________    _____    ______

      03-Jan-2018    10.59     72.47
      16-Jan-2018    11.02     67.98
      28-Dec-2017    23.75       776
      21-Dec-2017    15.74    120.88
      21-Dec-2017    12.02    125.51

then:

FinalSummed =

    4×3 table

         Date          X        Y   
      ___________    _____    ______

      21-Dec-2017    27.76    246.39
      28-Dec-2017    23.75       776
      03-Jan-2018    10.59     72.47
      16-Jan-2018    11.02     67.98

[KPSil]

Thank you this works well with my data however, the z values are summed within their dates and I would like the values to be summed chronologically as i.e date1=date1, date2=date1+date2, date3=date1+date2+date etc.

Do you have any advice on how to do this?

[Star Strider]

I would just use the cumsum function, for example:

FinalSummed = struct2table(FinalSummed)
FinalSummed.X = cumsum(FinalSummed.X)
FinalSummed.Y = cumsum(FinalSummed.Y)

This works, and appears to produce the correct result. Those additional lines go after the code I posted earlier (and the reason I included the struct2table call for reference).

[KPSil]

When I plot the data I only get a graph with the months on not the individual days within the months.

How do I specify the days?

[Star Strider]

See the documentation on xticklabels (link) and xtickformat (link). You might also need the xtickangle (link) function to rotate them to be readable.

In my experience, the axis date function results have proven to be difficult to modify. You will likely have to experiment to get the result you want.

Answer 2

[Peter Perkins]

In R2016b or later, timetables make the "compute daily sum" step super easy:

tt2 = retime(tt,'daily','sum')