Time-Series Forecasting With TimescaleDB and Prophet

This project combines the robust storage capabilities of TimescaleDB with the predictive power of Prophet for effective time-series forecasting. By leveraging TimescaleDB, users can efficiently store and manage large time-series datasets, while Prophet allows for accurate forecasting with its intuitive and powerful modeling capabilities.

Installation

Setting Up TimescaleDB

Install PostgreSQL and create a database.

Create a Timescale Instance.

Create a table named sales and convert it into Hypertable.

Populate the table with the given data.

Setting Up Python Environment

Ensure Python 3.8 or newer is installed on your system.

Navigate to the project directory and install the required Python libraries.

Usage

Preparing Your Dataset

Import your time-series data into TimescaleDB. Ensure your data is in a suitable format for time-series analysis.

Create a hypertable for your dataset using TimescaleDB to enable efficient querying.

Forecasting with Prophet

Use the provided Python scripts to extract data from TimescaleDB.

Run the sales prediction scripts to make predictions on your time-series data.

Visualize the data using Plotly to aggregate for different Time zones.