aggregating profiles of data

Question

Ashehad Ali 2015-6-11

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/223365-aggregating-profiles-of-data

评论： Ashehad Ali 2015-6-24

Hi there,

I would like to aggregate these three profiles of data-sets (HA01.csv, HA02.csv, HA03.csv) in matlab by determining the means so that the "new data-set" will have the same structure (i.e. the no. of rows and no. of columns as either of the three profiles and that the "new data-set" would contain means for the variables: SW1,SW2,SW3,SW4,SW5,SW6,SW7. That is, in the new-data set, for Hr=0, SW1= mean(x1,x2,x3), where x1 is the SW1 entry in HA01.csv when Hr=0, x2 is the SW1 entry in HA02.csv when Hr=0, where x3 is the SW1 entry in HA03.csv when Hr=0. The unknown values are reported in the data as -9999 or -9998 and when reporting the means, if at least one value is either -9999 or -9998, then the mean can be either -9999 or -9998.

I'm attaching the three data files. Would appreciate your feedback.

Ashehad

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Aarti Ghatkesar 2015-6-16

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/223365-aggregating-profiles-of-data#answer_182857

在 MATLAB Online 中打开

Hi Asheshad

The following code would do the operation of finding the mean for the columns SW1 to SW7 and write to a new file named 'newfile.csv'. Note that 'csvread' or 'dlmread' cannot write column headers.You can use the submission here to write column headers to a csvfile

numofcols=10;
numofrows=25;
ID=csvread('HA01.csv',1,0,[1,0,24,0])
Num=csvread('HA01.csv',1,1,[1,1,24,1])
Hr=csvread('HA01.csv',1,2,[1,2,24,2])
Final=[ID,Num,Hr];
for i=4:1:numofcols
     SW_file1=csvread('HA01.csv',1,i-1,[1,i-1,numofrows-1,i-1])
     SW_file2=csvread('HA02.csv',1,i-1,[1,i-1,numofrows-1,i-1])
     SW_file3=csvread('HA03.csv',1,i-1,[1,i-1,numofrows-1,i-1])
     SW=(SW_file1+SW_file2+SW_file3)./3
     Final=[Final,SW];  
end
csvwrite('newfile.csv',Final,1,0)