How to create a pivot table from this table (revised)

Suppose I have a table T

                         location      gender         
-----------------------------------------------------------
Customer 1       NY            male
Customer 2       LA            female
Customer 3       Austin       female
Customer 4       LA             female

Then I want to create a pivot table

                     Male         Female
--------------------------------------------
NY                1               0
LA                0               2
Austin            0               1

I checked unstack and grpstat but I am not sure. What will be a next step?

Answer 1

[Dave B]

Maybe there's a simpler path, but how about using groupsummary followed by unstack?

% Generate your table
cust = {'Customer 1' 'Customer 2' 'Customer 3'};
location = categorical({'NY' 'LA' 'Austin'},{'NY' 'LA' 'Austin'})';
gender = categorical({'male' 'female' 'female'},{'male' 'female'})';
t=table(location,gender);

% groupsumamry for counts, unstack for pivot, and then set NaNs to 0
s=groupsummary(t,{'location' 'gender'});
result=unstack(s,'GroupCount','gender');
result.female(isnan(result.female))=0;
result.male(isnan(result.male))=0;
disp(result)

    location    male    female
    ________    ____    ______

     NY          1        0   
     LA          0        1   
     Austin      0        1   

Answer 2

[the cyclist]

What you want to do is not really a pivot:

% Create your table
customer = {'1';'2';'3'};
location = {'NY';'LA';'Austin'};
gender = {'male';'female';'female'};
tbl = table(customer,location,gender)

tbl = 3×3 table
    customer     location       gender  
    ________    __________    __________

     {'1'}      {'NY'    }    {'male'  }
     {'2'}      {'LA'    }    {'female'}
     {'3'}      {'Austin'}    {'female'}

% You don't need the above. It is just to create your data.

% Create columns for male and female
tbl.male   = double(strcmp(tbl.gender,'male'));
tbl.female = double(strcmp(tbl.gender,'female'));

% Drop the gender column
tbl.gender = []

tbl = 3×4 table
    customer     location     male    female
    ________    __________    ____    ______

     {'1'}      {'NY'    }     1        0   
     {'2'}      {'LA'    }     0        1   
     {'3'}      {'Austin'}     0        1   

[alpedhuez]

Thank you. But actually I wanted a general method that allows

                         location      height       
-----------------------------------------------------------
Customer 1       NY            short
Customer 2       LA            tall
Customer 3       Austin       tall
Customer  4      LA            tall

with the output

  location    short   tall
    ________    ____    ______

     NY          1        0   
     LA          0        2 
     Austin      0        1  

[the cyclist]

Ah. I thought you just wanted to replace male/female with binary variables.

[alpedhuez]

Yes I should have provided more general examples. Thank you.