Figure 1
From: Automated Bayesian model development for frequency detection in biological time series
Effects of noise on precision. The effect of noise on σ for FFT (A) and BSA (B). The time series were simulated from d(t) = sin(ωt) + e, with ω = 0.5 rad/s and sampled with 3 s intervals to give 100 points. The noise level, e, varies between 0 and 100%. Although the qualitative behaviour of noise on the precision is the same for FFT and BSA in this example, BSA produces values of σ that are two orders of magnitude below the FFT values.