Macroeconomics dashboard

Macroeconomics dashboard is a python-based web application that downloads, processes and visualizes main macroeconomics indicators. The data is downloaded from World Bank and Organisation for Economic Co-operation and Development (OECD) via API. Downloaded data may be stored both in common csv files and in PostgreSQL database. Macroeconomics indicators are shown in the form of lineplots, histograms, maps, etc in the dashboard. Moreover, dashboard provides opportunity to use local LLM (via Ollama) for generating textual description of the current statement of global economy. Furthermore, plots may be saved in popular formats (png, jpeg, pdf, etc).

Illustrations

Technologies

• Data Processing: Pandas, NumPy
• Database: PostgreSQL, SQLAlchemy, psycopg2
• Visualizations: Plotly Dash
• LLM: Ollama

Installation

(recommended to use python virtual environment for managing all the dependencies of the project, following installation procedure implies that virtual environment is activated)

1. Clone the repository:

   git clone https://github.com/alexveider1/Macroeconomics-Dashboard.git
   cd Macroeconomics-Dashboard

2. pip install -r requirements.txt

3. launch with default params:

   python app.py

Configuration

Application has a lot of params for more flexible configuration. For setting param to its default value you do not need to add it to the launching command (just skip it)

• --preload_data: if you have launched the application before you can skip the step of downloading data by setting --preload_data to true (default false)
• --use_postgres: if you have PostgreSQL Server on your machine you may use it for storing data by setting --use_postgres to true (default false $-$ use csv-files for storing data)
• --postgres_username: PostgreSQL username required if --use_postgres set to true (default none)
• --postgres_password: PostgreSQL password required if --use_postgres set to true (default none)
• --postgres_port: PostgreSQL port required if --use_postgres set to true (default none)
• --use_ollama: if you have ollama installed on your machine you may use llm for textual description of the economics by setting --use_ollama to true (default false)
• --ollama_model: ollama llm model name (for example, llama3.2:1b or deepseek-r1:7b $-$ it is recommended to use small models for application acceleration) required if --use_ollama set to true, for example, (default none)
• --web_ui_port: port to use when launching web dashboard (default 0088)

For example,

python app.py --use_postgres true --postgres_username your_username --postgres_password your_password --postgres_port 5432 --use_ollama true --ollama_model llama3.2:1b