- #1
iibewegung
- 16
- 0
Hi,
Is there like some widely accepted threshold for autocorrelation time of a time series for it to be considered "uncorrelated"?
I used MATLAB to generate approx. 1000 gaussian random numbers and found their autocorrelation time to be approx. 1.07... is this small enough for it to be called white noise?
If not, can anyone tell me some ways to reduce the autocorrelation of a simple time series?
I heard sub-sampling is one way... is there anything else?
If someone can perhaps tell me the name of the method, I think I can look it up and do the rest.
Thanks in advance.
Is there like some widely accepted threshold for autocorrelation time of a time series for it to be considered "uncorrelated"?
I used MATLAB to generate approx. 1000 gaussian random numbers and found their autocorrelation time to be approx. 1.07... is this small enough for it to be called white noise?
If not, can anyone tell me some ways to reduce the autocorrelation of a simple time series?
I heard sub-sampling is one way... is there anything else?
If someone can perhaps tell me the name of the method, I think I can look it up and do the rest.
Thanks in advance.