Integer or discrete-valued time series are known to occur in many real-life applications. One of the major examples of such series is the very recent COVID-19 time series for the infected and death cases. There are multiple challenges associated in the modeling of the Integer-valued time series as such series exhibit high over-dispersion with several dynamic spikes. Besides, the time series processes need to be considered, that is, the auto-regressive or moving-average or the ARIMA models. The modeling also comprises of the inferential estimation procedures that need to be investigated. Usually, the conditional maximum likelihood seems to be the most plausible but may be computationally cumbersome. Alternatives such as the conditional least squares and method of moments must be considered. The course and workshop aim to bring together researchers, students, and practitioners of time series and forecasting, to present and discuss new advances in the field, related to theoretical, methodological developments or novel applications. This course and workshop offer also a detailed background on the integer-valued time series models and the different thinning mechanisms to establish the relationship between the successive observations. In addition, the construction of the autoregressive and moving average models is also considered. The computational language R is used to simulate the different integer-valued series based on the different structures and to implement the inferential procedures. Data examples on COVID-19, crime, and financial series are referred to validate the models and the forecasts.