Kerr hypothesis test using the GLM method applied to the n = 1 subring. An image was simulated for a BH of spin a = 0.94 and inclination i = 17°, with equatorial emission profile parameters μ = r₋, , σ = 0.5. (Upper left) Amplitude |V(u, φ)| of the visibility at φ = 135°, with baselines colored according to which subring dominates the signal (Sec. 2.7). (Upper Right) Visibility amplitude fit (with envelope) in the n = 1-dominated baseline window u_w ∈ [30, 70] Gλ (Sec. 3). (Lower left) Goodness of fit (Eq. (36)) as a function of fitting diameter . The dot corresponds to the diameter with maximal goodness of fit . The error bar includes nearby diameters such that RMSD_u(d_φ) ≤ 2 RMSD_u,min, or equivalently . (Lower Right) Fitting to a circlipse using the multi-fit method (Sec. 3.2.4), with the same prescription for the error bars. Angles φ = 125° and 130° were removed because the visamp fit in the chosen baseline window was too poor.