search

team-sagil / TimeDelay

Final project for the AstroML course, based on data from TDC1
0 stars 1 forks source link

autocorrelation #3

Closed asadisaghar closed 8 years ago

asadisaghar commented 8 years ago

why does the autocorrelation of the two signal within each sampling window is EXACTLY the same for both light curves?

asadisaghar commented 8 years ago

because correlate apparently identifies the end of each window as the period of the signal. so the maxcorr for autocorrelation is always (time[window][-1] - time[window][0]) / 2

redhog commented 8 years ago

How do you avoid this / how can we make it not identify that as the signal period? Is it unable to identify signals with a longer period than the sample window?

asadisaghar commented 8 years ago

I don't know. But this is not one of the things we want to measure.