Packet video and voice are %is the biggest traffic sources in todays Internet. Numerous papers have been written about the influence of packet loss on video and voice quality. This tutorial is motivated by a series of papers \cite{MarkUnd}, \cite{Markovich2013-2}, \cite{Markovich2013-3}, \cite{Markovich2013}. In these papers, we try to understand causes of packet loss during packets transmission through Internet and how that affects the quality of transmitted video and voice data. To support an adequate teletraffic analysis of the packet traces by the evaluation of relevant performance indices, for instance, of captured packet flows arising from new real-time services in a multimedia Internet, advanced statistical methods are required. They allow us to cope with immanent dependencies and underlying heavy-tailed distributions of interesting features of the traffic such as the bitrates, the inter-arrival times, bit loss and delay distributions of the packet streams, the distribution of path length of the packet, the probability to loss packets or the equivalent bandwidth as a control parameter. In the tutorial we shall discuss useful statistical and probabilistic techniques to handle the arising strongly correlated or long-range dependent time series and heavy-tailed marginal distributions that determine the underlying random variables of the observed data features. New procedures to compute the demanded bandwidth of observed streams or the delay-loss profiles of packet flows during a session will be stated. The inferences are partly based on the last achievements in the theory of extremes in time series. The analysis concepts will be illustrated by real traces arising from some popular Internet applications. The tutorial shall stimulate the participants to incorporate adaptations of the sketched procedures into the practice according to their personal needs.