How to compute the time lag by cross-correlation of ir-regular time series, accurately?

HI everyone,

My data set consists of monthly observations of two parameters (already normalized) over a 50 years time window?

I require to compute the time-lag between these two time series on yearly basis, at first. IN addition, I also need to compute the time lag over 5 years windows or any other window length.

May someone suggest how i can accurately compute the phase lag.

Here is what I did so far:

% Data presentation 

cd_ev=readmatrix('file.csv'); % selected candidate earthquake
tim=datenum(cd_ev(:,3),cd_ev(:,4),cd_ev(:,5),cd_ev(:,6),cd_ev(:,7),cd_ev(:,8));
p1=cd_ev(:, 9);
p2=cd_ev(:, 10);
figure(1)
plot(tim, p1)
hold on
plot(tim, p2)
hold off

Cross-correlation result

[c,lags] = xcorr(p1,p2);
[c,lags] = xcorr(p1,p2,'normalized');
figure(2)
stem(lags,c)

Thank you!

[KSSV]

You may get botht the time series into same time stamp using interp1 and than you may try using crosscorr.

[Andi]

@KSSV I already scale them at same interval one observation per month.

[KSSV]

Then try crosscorr