How can i compute the mean of specific number of table rows?

Asked by Ioannis Tsikriteas on 9 Aug 2017

Latest activity Commented on by Cong Ba on 10 Aug 2017

3 views (last 30 days)

Hi, I have a large amount of table data which is divided by date and time....to be more specific i have data that is per 10 minutes and i want to compute the mean of them per hour! In other words, i want ot compute the mean of my data per 10 rows until my data is finished! Is this possible and if yes....which command should i use?

0 Comments

3 Answers

Link

Answer by Cong Ba on 9 Aug 2017

Accepted Answer

Try reshape:

a = randn(100,1);  % assume this is the column you have
b = reshape(a,[10,10]); % reshape it so each column has 10 rows
avg = mean(b);

4 Comments

Show 1 older comment

Cong Ba on 9 Aug 2017

let's assume you throw the first 2 rows...

a = randn(278984,2);  % assume this is your table
a = a(3:end,:); % assme throw out the first two rows
b1 = reshape(a(:,1),[6, 278982/6]); % reshape the first column to [6, 278982/6] so it has 6 rows
avg1 = mean(b1);
b2 = reshape(a(:,2),[6, 278982/6]); % same to the second column
avg2 = mean(b2);

avg1 would be the hourly series you want for the first column of your data. avg2 would be for the second.

Ioannis Tsikriteas on 10 Aug 2017

When i try the reshape command i get the following message: Error using tabular/reshape (line 150) Undefined function 'reshape' for input arguments of type 'table'.

To be more clear isend a pic of the shape of my data tables

Cong Ba on 10 Aug 2017

Andrei is using the timetable function which may better suit your need (you essentially have a time series). But if you want to manipulate data in matrix form (my code works for matrix form), you can try this function: table2array

Link

Answer by Andrei Bobrov on 9 Aug 2017

n = 278984;
t = minutes(10*(1:n)');
D = randi(255,n,2); % Let D - your data (278984 x 2)
TT = timetable(t,D(:,1),D(:,2));
TT_out = retime(TT,'hourly','mean');

2 Comments

Ioannis Tsikriteas on 10 Aug 2017

I am sorry but i didn't understand the concept of these commands.

Mine data are already ona a table 278984x2 and has the following shape:

I am searching a way to compute the mean between the fifth and the eleventh row for example in order to have my data per hour, not per 10 minutes

Andrei Bobrov on 10 Aug 2017

a.x_10_double = [a.x_10_{:}]';
TT = table2timetable(a(:,[1,3]),'RowTimes','x08_Jul_100_10_00');
TT_out = retime(TT,'hourly','mean');

a2 = table(a{:,1},[a{:,2}{:}]','v',{'datetime','x_10_double'})
a2.groups = hour(a2{:,1});
a_out = varfun(@mean,a2,'Group','groups');

Link

Answer by Peter Perkins on 10 Aug 2017

Add a grouping variable to your table and use varfun. Something like

n = ceil(height(t)/10);
g = repelem(1:n,10)';
t.Group = g(1:height(t));
t.groupMeans = varfun(@mean,t,'GroupingVariable','Group')

You are now following this question

Ioannis Tsikriteas (view profile)