We consider the estimation of the affine parameter (and power-law exponent)in the preferential attachment model with random initial degrees. We derive thelikelihood, and show that the maximum likelihood estimator (MLE) isasymptotically normal and efficient. We also propose a quasi-maximum-likelihoodestimator (QMLE) to overcome the MLE's dependence on the history of the initialdegrees. To demonstrate the power of our idea, we present numericalsimulations.
展开▼
