Convert table of Daily Stock data into Monthly data

I have a 13000x3000 table with colums representing the dates and rows representing the stocks with values of daily stock returns in it.

Now I'd like to get the monthly sum of returns for each stock (each line).

Would appreciate any kind of help.

[Housam]

Answer = sum (Data,2)

https://www.mathworks.com/help/matlab/ref/sum.html

Answer 1

[Cris LaPierre]

Look into groupsummary.

G = groupsummary(T,groupvars,groupbins,method)

Use the date info as your groupvar.

Use the appropriate value for your groupbin (sounds like you want 'monthly').

Specify the desired method (sounds like you want 'sum')

[Cris LaPierre]

Just a comment that groupsummary expects your data to be arranged in columns/vectors. It looks like you will have to transpose your data so that the dates correspond to rows, not columns.

[Fabian Niederreiter]

Thank you Cris. I agree that groupsummary might be the way to go here. Nevertheless I still get a few errors that I don't really know how to interpret. Maybe you can help again.

My code:

Error messages:

[Cris LaPierre]

For more help, you should share your data. Save your variables to a mat file and attach them to your post using the paperclip icon.

[Fabian Niederreiter]

The data unfortunatelly exceeds the 5MB limit by far (even as a zip).

I did attach my current code tough. Maybe you can spot the mistake through that. Would be really happy about some more feedback.

[Cris LaPierre]

It would appear that your grouping variable, dates, is not valid with the syntax you are using.

B = groupsummary(A,groupvars,groupbins,method)

When you use the table form, your grouping variable needs to be a variable in the table. Try adding dates as a new column in your table rather than as the rownames.

G_CRSP.dates = dates;

Then your groupsummary statement should work.

[Fabian Niederreiter]

Thanks for your help Chris! With some adjustments to my data I managed to make it work with the help of groupsummary. :)