Nowcasting and Forecasting with Text as Data

This repository contains teaching materials — including slides, code, and datasets — for learning how to build text-based indicators and apply them in forecasting and nowcasting tasks. Through practical, hands-on examples, it covers zero- and few-shot learning for text classification, event detection with limited labels, and mixed-frequency data techniques for time series modeling.

📦 Repository Structure

BSE-FORECASTNLP/
├── session1/
├── session2/
├── session3/
├── .gitignore
├── README.md
└── requirements.txt

• Session 1: Zero- and Few-Shot Learning for Text Classification
• Session 2: Nowcasting Political Events with Limited Labeled Data
• Session 3: Mixed Data Sampling Approaches for Time Series Nowcasting and Forecasting

✅ Setup Instructions

1. Ensure Python 3.11 is installed:

   Before proceeding, verify that Python 3.11 is installed:

   python3.11 --version

   If not, download and install it from the official Python website.

2. Clone the repository:

   git clone https://github.com/RenatoVassallo/BSE-ForecastNLP.git
   cd BSE-ForecastNLP

3. Create a virtual environment with Python 3.11:

   python3.11 -m venv .venv
   source .venv/bin/activate   # On Windows: .\.venv\Scripts\activate

4. Install the required libraries using the latest requirements.txt:

   Ensure that you only use the requirements.txt file from the GitHub repository. Avoid using any previous versions.

   pip install -r requirements.txt

5. You are ready to go!

   Navigate to the session1 folder and try running any notebook.

Feel free to open issues or reach out if you have any questions. 🚀