Calculating weekly mean from daily data.

I have a netcdf file with u and v components for wind. I want to calculate the weekly averages of the u and v components based off of daily data.

so far I have

[y,mo,d] = datevec(dn);
xm = zeros(120,1); 
for i = 1:120
   if d = 1:7
   xm(i) = mean(v_vector(d == i));
   elseif d = 8:14
       xm(i) = mean(v_vector(d == i));
   elseif d = 15:23
       xm(i) = mean(v_vector(d == i));
   elseif d = 24:21
       xm(i) = mean(v_vector(d == i));
end

dn is the date number. where 1 to 120 is the time increments from 1-Jan-2009 to 31-April-2009.

Thanks in advance

Answer 1

[Eric Lin]

First, I would recommend finding a reliable method of determining the week number of a particular date. If you have access to R2014b or later, you can use the week function. An alternative is also available on MATLAB File Exchange.

From there, you can use a loop and logical indexing to find the means of each week:

%sample dates
t = datetime(2015,05,31):datetime(2015,06,18)

%week numbers
w = week(t)

%sample data
data = randi(10,1,19)

%mean for week 24
mean_24 = mean(data(w == 24))