Looks like a few frequencies fitted, some Fourier components,...

Looks like a few frequencies fitted, some Fourier components, envelope whose top then corresponds to SPX top.

From memory and trying to visualise it without putting pen to paper, if some of the dominant frequencies are closely spaced, then envelope has a cycle itself with much long wavelength. So can in effect at some, but not all points unless full Fourier is done, fit some of the detail as well as long term character.