#45 A long-short Bayesian State-Space model with R and RStan

Welcome to our 2nd own Berlin RUG online meetup! We'll learn about bayesian modeling for a forecast on our upcoming election. And we will get to see each other again, so we're hoping many will sign up :)

A long-short Bayesian State-Space model for the German Federal Election 2021 with R and RStan

# Description
Marcus Groß and Michelle Golchert will present a Bayesian model for forecasting the German federal election 2021 with R and RStan that gives more insights and more accurate predictions compared to simple poll-averaging methods. The model consists of two parts:
A model for the election outcome (i.e. the vote share for each party based on poll data) and a model to assign a probability to the coalition of the actual government using expert survey data. An in-depth description and motivation is discussed in this talk. The model is fitted with (R)Stan, a Bayesian modeling framework, that can seamlessly be integrated into R.

# Agenda
18:30 Intro
18:35 Talk
19:05 Q&A
19:30 Closing remarks (we can extend it if needed)

# Bios
Marcus Groß and Michelle Golchert are Data Scientists at INWT Statistics.
Marcus holds a Ph.D. in Statistics and has several years experience consulting on and managing projects in the data science field. He is an expert in predictive analytics and Bayesian statistics.
Michelle has a master's degree in psychology and specializes in supporting Data Science projects in the areas of software development, DevOps and CI.