Speakers:
Forecasting Public Expenditure (BAföG) Using LSTMs: A Multi-Task Learning Approach for Volatile Data
Date:
Tuesday, November 18, 2025
Time:
2:35 pm
Summary:
Accurately forecasting expenditures is crucial for budgeting and policy planning. This session presents a multi-task LSTM model designed to predict annual federal student aid (BAföG) expenditures for the current and next two years. Using 16 state-level time series, it addresses extreme intra-year volatility, strong seasonality, and the risk of overfitting with small datasets. Attendees will gain insights into overcoming these challenges and improving real-world time series forecasting models.
Speakers:
Who Will Be in the Bundestag? - A Bayesian Approach to Election Forecasting in Germany
Date:
Tuesday, November 18, 2025
Time:
3:05 pm
Summary:
Traditional Sunday polls offer vote snapshots, but INWT Statistics’ election forecast project provides deeper electoral insights. This case study details their Bayesian state-space model, which is based on polling data and historical election results. While their 2025 vote share predictions were on par with simple poll averages, the model’s true power lies in calculating probabilities for crucial events like coalition majorities or parties clearing the 5% hurdle.