Finding the mean at each interval

Hi everyone,

I have data containing 2 columns; Temperature and V. Temp contain repititive temperature values and V contains (2,3 and 4). What I want is that for each 1°C temperature I want the mean of values in V. For example at 12°C, a number of values (2,2,3,4,4,2,2) fall in this inteval (12°C), so, therfore at 12°C = 2.7. Any suggestions how to do this? Data is attached. Thank you

Answer 1

[KSSV]

T = readtable('https://in.mathworks.com/matlabcentral/answers/uploaded_files/746119/Data.xlsx') ;
Temp = T.Temp ; 
V = T.V ; 

t0 = 10; t1 = 12 ;   % Intervel 

idx = Temp > t0 & Temp < t1 ;

iwant = mean(V(idx))

iwant = 2.6726

[Zhou Ci]

Why is it giving just 3 values? There is alot of repitition of temp values, I want output to be like this;

For example,

at -22°C mean of V to be 3;

-15°C: 2.5

-10°C: 2.8

8°C: 2

29°C: 2

and so on

[KSSV]

Edited the code

[Zhou Ci]

I want values (a complete column vector at each temp interval): Minimun value of temp is -72.95 and maximum is 37.45.

So at each temp interval I want a mean of V.

Answer 2

[Mathieu NOE]

hello

my 2 cents suggestion - see if for the empty intervals you prefer naN output (red dots) or you prefer interpolated data (black dots)

plot

code :

clc
clearvars

Tab = readtable('Data.xlsx') ;
T = Tab.Temp ; 
V = Tab.V ;

[Ts,ind] = sort(T); % sort T in ascending order
Vs = V(ind);

minT = floor(min(Ts));
maxT = ceil(max(Ts));

Tnew = minT-0.5:1:maxT+0.5; % interval like -10.5 to 10.5 
Tplot = minT:1:maxT; % corresponding center value (10)

for ci =1:length(Tnew)-1
    ind = find(Ts>=Tnew(ci) & Ts<=Tnew(ci+1));
    
    if ~isempty(ind)
        Vmean(ci) = mean(Vs(ind));
    else
       Vmean(ci) = NaN; 
    end
end

% optionnal : remove NaN (missing) values by pchip interpolation
Vmean2 = interp1(Tplot,Vmean,Tplot,'pchip');

figure(1)
plot(Ts,Vs,'b',Tplot,Vmean,'-dr',Tplot,Vmean2,'*k');
legend('raw data (sorted)','averaged (including NaNs)','averaged (interpolated)');