Statistical error in f_NL as a function of the size of the prior on p, for three values for p, shown in different colours, as indicated in the legend. The mock data vector for this figure is generated using the linear power spectrum computed with CAMB in Eq. (11) and fixing b₁ = 1.4 and f_NL,  fid = 0. The horizontal dashed lines show the case where we performed the fits fixing p in the model to its corresponding fiducial value. The dashed vertical line displays the value of the error in p, obtained for halos with mass between 5 × 10¹¹ and 1 × 10¹² h⁻¹ M_⊙ (bin 4 in Table 3).