for analysis of timeseries signal to attempt to find and classify periodic signals, except unlike Fourier Transform approaches, it works in an unevenely sampled/missing data environment. In essence, use like Discrete Fourier Transform/Fourier Transform in cases where observations are unevenely spaced or noisy.

  • means it can handle irregular sampling and wont accumulate bias by interpolating missing data.
  • equivelant to a least squares fit of sine/cosine functions to data.
  • all frequencies from lombscargle are not angular frequencies, rather frequencies of oscillation (num cycles per unit time)