Least squares method is not appropriate for estimating 16304551&_mathId=si63.gif&_user=111111111&_pii=S0378437116304551&_rdoc=1&_issn=03784371&md5=44be2b539ba87f104e9f0d53ed05bdf5">16304551-si63.gif">$\beta$ of Barabási–Albert graphs.

A modified version of Clauset et al. (2009) is more preferable for estimation.

Simulated graphs show that exponent is not the theoretical 16304551&_mathId=si64.gif&_user=111111111&_pii=S0378437116304551&_rdoc=1&_issn=03784371&md5=ad32e01cf5125f1d3c4c81128b63c604">16304551-si64.gif">$\mathrm{\beta \; =\; 2}$ for small graphs.

Value of 16304551&_mathId=si63.gif&_user=111111111&_pii=S0378437116304551&_rdoc=1&_issn=03784371&md5=44be2b539ba87f104e9f0d53ed05bdf5">16304551-si63.gif">$\beta$ is also dependent on number of edges added, 16304551&_mathId=si21.gif&_user=111111111&_pii=S0378437116304551&_rdoc=1&_issn=03784371&md5=c7e82b1375e888385c12dfdbc76f3214" title="Click to view the MathML source">m$m$, for small graphs.

Smaller value of m results in larger value of 16304551&_mathId=si63.gif&_user=111111111&_pii=S0378437116304551&_rdoc=1&_issn=03784371&md5=44be2b539ba87f104e9f0d53ed05bdf5">16304551-si63.gif">$\beta$ for all graph sizes.