Talk:Whittle likelihood

Latest comment: 2 years ago by 2A02:908:1085:FA60:407B:2CDD:1BD1:D8CB

Hi. I think there is a factor of 2 missing somewhere in the "Definition" section: if S1 is the one-sided power spectrum of X(t), then it, times 1/2, is the variance of its Fourier transform squared. Since that variance is equal to the sum of the variances of the real and imaginary parts of the Fourier-transformed variable, then the variance of each of them must be 1/4 the one-sided power spectrum, not the power spectrum as stated? 2A02:908:1085:FA60:407B:2CDD:1BD1:D8CB (talk) 20:04, 5 November 2021 (UTC)Reply