Calculate mean from daily to monthly values

I have I have a data containing daily pricipitation data from 2001 to 2010, each year I have 1 file

lon=ncread(ncfile,'lon');
lat=ncread(ncfile,'lat');
precip=ncread(ncfile,'precip');

precip 360 x 280 x 365
lat 280x1
lon 360x1

I want to calculate the monthly mean for each year and then calculate average for 10 years.

the slow way I can do like that:

JAN{K}=nansum(precip{K}(:,:,1:31),3); %K is number of year
FEB{K}=nansum(precip{K}(:,:,32:59),3);
JAN=cat(3,JAN{:});
JAN=mean(JAN,3);

but for 2004, 2008 we have 366 day and this way is not very good to calculate

How can I do that with loop?

Thanks

Answer 1

[convert_to_metric]

Hi minh lan,

You can consider making use of the month funciton. Take a look at the following code, perhaps it will lead you to a solution.

data=[1:365]*10; % just some example data
length_of_precip_data=length(data); % you might need to use the size function in place of length depending on how your data is organized
year=2010; % for example
start=datetime(year,1,1); 
m=month(start+caldays(1:length_of_precip_data)-1); % now you have an array the same size as your data that indicates which month an element is in

% so if you want to find just values from february:
data(m==2) % february is the second month, ie m==2

Answer 2

[Steven Lord]

Use the groupsummary function or store your data in a timetable and use retime on that timetable.

Answer 3

[ABHILASH SINGH]

You can try this;

https://in.mathworks.com/matlabcentral/fileexchange/74497-daily-data-to-monthly-average-sum-in-matlab?s_tid=LandingPageTabfx