Galaxy stellar mass vs. host halo mass at z = 0 for the set of parameters (Table 1) that fit best the galaxy stellar mass function. Each point is one of our simulated galaxies. The symbol sizes and colours vary with the fraction f_merg of mass acquired in mergers, and the fraction g_cold of this mass acquired by cold-mode mergers (where the sum of the masses of the haloes of the merging galaxies is below M_shock), respectively, according to the legend. The vertical green line corresponds to M_halo = M_shock, while the oblique magenta line indicates m_stars = f_b   M_halo. The green curves show the stellar masses of galaxies acquired through the accretion channel (Eq. (8) with best-fit parameters from Table 1) at z = 0 (lower) and z = 3 (upper curve). The shaded cyan and gold regions correspond to the observations from the SDSS (Yang et al. 2009), and the halo-galaxy abundance matching of Guo et al. (2010). For the latter, the halo masses were shifted by log ₁₀h to conform to our units. Also, the stellar masses were shifted by  − log ₁₀h and log ₁₀h for Yang et al. and Guo et al., respectively, to take into account the different units with h. We also shifted the stellar masses by an additional factor of +0.1 and  + 0.15 dex to pass from the Kroupa and Chabrier IMFs, respectively assumed by Yang et al. and Guo et al. to the diet Salpeter IMF assumed by Bell et al. (2003), on whose mass function we have calibrated our model.