The amount of video traffic on the Internet has seen a tremendous increase over the past few years. In 2020, it is predicted to account for 85% of the total Internet consumer traffic. Due to this dominant role, streaming traffic has to be considered by workload models used to evaluate the performance of networking systems. A de facto standard technology for Internet-based Video on Demand (VoD) services is HTTP-Based Adaptive Streaming (HAS), which is also increasingly used for live streaming. Unfortunately, HAS clients produce a very specific workload pattern that is not appropriately represented by traditional HTTP traffic models. In the present work, we propose a stochastic model accurately describing such traffic, along with a methodology to generate synthetic traffic using functionality provided by commonly available numerical/scientific software libraries. We perform a proof of concept by fitting the model to a data set collected in a residential Wi-Fi environment, and generating synthetic traffic matching the characteristics of the traffic in the collected data set.