684 10 archive 91485 disk0/00/00/06/84 2026-04-08 16:46:59 2026-04-08 16:46:59 2026-04-08 16:46:59 dataset show Neunhoeffer Marcel 0000-0002-9137-5785 Gschwend Thomas 0000-0002-8656-9622 Müller Klara 0000-0001-9314-4094 Munzert Simon Stoetzer Lukas 0000-0003-2486-7496 320 10320 60700 We present the Zweitstimme Model to forecast the outcome of the 2021 German federal election. The dynamic Bayesian forecasting model combines data from published pre-election public opinion polls with information from fundamentals-based forecasting models. The model takes care of the multiparty nature of the setting and generates probability statements about other quantities of interest, such as the majority for coalitions in parliament, and the expected overall size of parliament. As a central expectation, 100 days before the election, our model predicts a strengthening in Green party support (17% [12%; 23%]) and new majorities for coalition governments (67% chance for a majority of a CDU/CSU-Green coalition). The model further expects the Bundestag to grow in size to around 814 seats. eng 2021 https://doi.org/10.7910/DVN/EDTKNW metadata cc0 Gschwend, Thomas und Müller, Klara und Munzert, Simon und Neunhoeffer, Marcel und Stoetzer, Lukas F. (2022), <a href='https://madoc.bib.uni-mannheim.de/id/eprint/61495' target='new'>The Zweitstimme model: A dynamic forecast of the 2021 German federal election</a> FALSE