Python for Time Series Forecasting

Welcome to the repository for the LinkedIn Learning course Python for Time Series Forecasting by Jesus Lopez. The full course is available on LinkedIn Learning.

Course Overview

Many business forecasts miss the mark, wasting millions through over- or under-planning. Without the right tools and a mental framework for understanding time series, predictions can easily lead to costly mistakes. In this course, you'll master forecasting with step-by-step guidance in Python, learning theory exactly when you need it so you can put it into practice.

You'll work with real datasets from the US Energy Administration and Federal Reserve, learning to:

• Load, pre-process, and visualize time series data
• Build and validate ARIMA, SARIMA, exponential smoothing, and Prophet models
• Use train/test split and walk-forward validation to trust your forecasts and plan for the future

Ready to forecast like a pro? Join Jesús López and turn raw data into future-proof forecasts.

Getting Started

Using Codespaces (Recommended)

This course is designed to be run in a GitHub Codespace—a virtual computer with all required Python packages pre-installed. To get started:

1. Click the Code button in GitHub and select Create codespace on main.
2. Wait for the Codespace to finish setting up. You can monitor progress by clicking the bell icon in the Codespace interface.
3. Once setup is complete, activate the VS Code Icons extension for a better folder view (optional).

Repository Structure

All course notebooks are in the notebooks folder, organized into three main blocks:

• 1_Foundation: Loading, pre-processing, and decomposition of time series data
• 2_Modelling: Building models (ARIMA, SARIMA, smoothing, Prophet), understanding stationarity, and evaluating models
• 3_Validation: Simulating real-world scenarios with rolling window validation

Inside each folder, you'll find Jupyter notebooks with code ready to run. To execute code cells, select a Python environment (e.g., global env) when prompted, then use Shift+Enter to run each cell.

Assignments

Each block contains an assignment notebook. These follow a template approach: substitute the UPPERCASE variables with your chosen data and follow the steps to validate your results and forecast into the future.

Requirements

All requirements are pre-installed in Codespaces. If running locally, install dependencies from requirements.txt.

Cloning the Repository

To work locally:

1. Clone this repository: git clone
2. Install requirements: pip install -r requirements.txt
3. Open the notebooks folder and start exploring!

---

Happy forecasting!