Time Series Analysis with Python Course Syllabus

Overview: This hands-on course guides you from time series fundamentals to forecasting and deployment, blending theory with practical coding exercises. You'll explore data manipulation, visualization, decomposition, classical models like ARIMA and Prophet, model evaluation, and introductory advanced techniques. With approximately 8 hours of total content, the course emphasizes real-world applications and code-first learning, preparing you to confidently build and deploy temporal models. Lifetime access ensures you can revisit concepts anytime.

Module 1: Introduction to Time Series Analysis

Estimated time: 1 hours

• Time series definitions and characteristics
• Understanding trend, seasonality, and cyclical components
• Common use cases in industry
• Loading and plotting basic time series data

Module 2: Data Manipulation & Preprocessing

Estimated time: 1.5 hours

• Date and time indexing with Pandas
• Resampling time series data
• Handling missing timestamps
• Rolling and window functions

Module 3: Visualization & Exploratory Analysis

Estimated time: 1 hours

• Creating line plots for temporal data
• Interpreting autocorrelation (ACF) and partial autocorrelation (PACF) plots
• Generating seasonal subplots
• Identifying patterns through visual inspection

Module 4: Decomposition & Stationarity Testing

Estimated time: 1.5 hours

• Additive vs. multiplicative decomposition
• Decomposing series into trend, seasonal, and residual components
• Performing Dickey–Fuller stationarity tests
• Applying detrending and differencing techniques

Module 5: Classical Forecasting Models (ARIMA/SARIMA)

Estimated time: 1.5 hours

• Understanding AR, MA, ARMA, and ARIMA models
• Identifying p, d, q parameters using ACF/PACF
• Building and tuning SARIMA models
• Using grid search for model order selection

Module 6: Exponential Smoothing & Prophet

Estimated time: 1 hours

• Simple and Holt’s linear exponential smoothing
• Holt–Winters seasonal smoothing
• Introduction to Facebook Prophet
• Comparing ETS and Prophet on seasonal datasets

Module 7: Model Evaluation & Tuning

Estimated time: 1 hours

• Train/test splits for time series
• Walk-forward validation strategy
• Computing MAE, RMSE, and MAPE
• Backtesting model performance

Module 8: Advanced Topics & Deployment

Estimated time: 1 hours

• Introduction to multivariate forecasting with VAR
• Basics of LSTM for time series
• Scheduling forecasts using cron or Airflow
• Containerizing forecasting scripts for deployment

Prerequisites

• Basic familiarity with Python programming
• Understanding of Pandas for data manipulation
• Fundamental knowledge of data analysis concepts

What You'll Be Able to Do After

• Analyze and visualize time series data effectively
• Preprocess and clean temporal datasets
• Build and tune classical forecasting models like ARIMA and Prophet
• Evaluate models using proper time series validation techniques
• Deploy forecasting pipelines in production-like environments