How to know the date to which a group (findgroups) corresponds?

Hello!

I have a timetable (TT) where one column is a Date (dd/mm/yyyy format) and another is an Hour, I applied the following code to classify them by groups:

groups_TT = findgroups(TT.Date, TT.Hour);

This gives me a unique number for each day and hour. My question is, how can I get an identifier from this number that tells me the Date it corresponds to? The unique number is not as useful for me to perform my analysis. I need to know the date that this number corresponds to.

I attach a small example. Here my code will give me a unique identifier (from 1 to 6) and I would like to create an additional column that tells me what date each one corresponds to (if possible another table with two columns: unique id and date)

Thanks in advance!

Angela

[KSSV]

Read about ismember.

Answer 1

[Adam Danz]

findgroups()

Use the 2nd output [G,ID] = findgroups(A)

G(i) belongs to ID(G(i),:). For example,

A = ["D" "B" "B" "A" "D" "D" "D"];
[G,ID] = findgroups(A)

%     G =
%          3     2     2     1     3     3     3
%     ID = 
%       1×3 string array
%         "A"    "B"    "D"

So, G(1) shows that A(1) is in group #3. Group #3 is ID(3) or ID(G(1)) which is "D".

To find all samples in your data that belong to a specific group,

isGroupMember = G==find(ID=="D");  % or strcmp(ID,"D")
% or,           G==3

isGroupMember is a logical vector the same size as A and contains 1s (True) where A=="D".

unique()

Note that [ID,~,G] = unique(A) produces the same outputs (but G is a column vector)!

Use [ID,~,G] = unique(A,'rows') when A is a matrix and the columns contain the grouping variables.

%     ID = 
%       1×3 string array
%         "A"    "B"    "D"
%     G =
%          3
%          2
%          2
%          1
%          3
%          3
%          3