How to get monthly average from daily data?

Question

frankovaT 2018-9-27

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/421053-how-to-get-monthly-average-from-daily-data

编辑： jonas 2018-9-27

data.xlsx

So i have this Matrix, consisting of 4 columns: Column 1: Year Column 2: Month Column 3: day (1 to 30) Column 4: Some Value

I want to make monthly average of "some value" without breaking the matrix into some new matrix and calculate mean for each moths matrix seperately then rejoin them. I want to make an average matrix using for loop and find function for each myear seperately. So I dont like to have an average from all the first months of all the years together. Does anyone have an Idea?

I have sth like this, but i donno why it is not working:

k=1
  %Calculating Monthly Average
  for i=1961:1990                                     %Years Looping
      for j=1:12                                      %Months Looping
          idx=find((date(:,1)==i) & (date(:,2)==j));  %Find all indeces for the specific month
          meandata1=mean(date(idx,4),1);
          Monthly_Ave(k,:)=[i,j,meandata1;];          %Store in a new matrix the mean values per month
          k=k+1;
      end
  end

5 个评论
显示 3更早的评论隐藏 3更早的评论

frankovaT 2018-9-27

Dear jonas I truly appreciate your help i attached an Excel file of my data.

jonas 2018-9-27

and answer updated as promised!

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

jonas 2018-9-27

3
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/421053-how-to-get-monthly-average-from-daily-data#answer_338715

编辑：jonas 2018-9-27

在 MATLAB Online 中打开

Not sure what you mean by "without breaking the matrix into some new matrix", but here we go:

x is column with months,

y is data

[G,name] = findgroups(x)
out = splitapply(@mean,y,G)

This will give you max 12 values, one for each month of the year. If you have several years, and you want 12 values per year, then you can do the following trick.

x1 is months (double)

x2 is years (double)

x=cellstr(num2str([x1, x2]))
[G,name] = findgroups(x)
out = splitapply(@mean,y,G)

Better yet, put your data in a timetable and use retime to calculate the monthly average. Here's an example using your data:

data=xlsread('data.xlsx')
t=datetime(data(:,1:3))
T=timetable(t,data(:,4))
T2=retime(T,'monthly','mean')

6 个评论
显示 4更早的评论隐藏 4更早的评论

jonas 2018-9-27

编辑：jonas 2018-9-27

在 MATLAB Online 中打开

There's a typo here:

Monthly_Ave(k,:)=[i,j,meandate1;];

should be meandata1. Also date is a built in function, so don't name your variable that! Both methods give the same results, so that's reassuring :)

frankovaT 2018-9-27

Oh god! you are an angle! I was so unsure about my code that i didnt even tried to look for typos:))) My first for loop!!! thanks :)

请先登录，再进行评论。

Answer 2

KSSV 2018-9-27

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/421053-how-to-get-monthly-average-from-daily-data#answer_338714

在 MATLAB Online 中打开

Let M be your month column, extracted and A be your data.

B = zeros(12,1) ;   % means 
for i = 1:12
    B(i) = mean(A(M==i)) ;
end

1 个评论
显示 -1更早的评论隐藏 -1更早的评论

frankovaT 2018-9-27

maybe I should have mentioned that i have 50 years of data, so if I ll have like 50, Marches, and this is going to make an average from all the Marches for instance. but i want to make an average form each daily vaule in month of a year seperately.

请先登录，再进行评论。