Time to First Citation Estimation in the Presence of Additional Information

We are interested in modelling the time to first citation, that is how long does it take for a publication to be cited for the first time after it has been published in a journal. We argue that both cited and uncited publications should contribute to the distribution of the time to first citation. Moreover, our objective is to model the time to first citation nonparametrically, hence under no parametric assumption. Due to the similarities with the observed data in survival analysis, we employ the techniques based on censored data and describe the distribution of the time to first citation in terms of the hazard rate, that is the instantaneous rate of being firstly cited. We find that publications receive their first citation at increasing rates in the first 24 months after their publication date and at decreasing rates afterwards. Moreover, we observe that the hazard rate and hence the time to first citation is influenced by the document type, number of authors and collaboration type and field. We also investigate the difference in the time to first citations for publications grouped by their collaborative status or the assigned field. Conference Topic Citation and co-citation analysis