Time Series Masterclass

A 24-hour masterclass on forecasting and anomaly detection — from classical methods (ARIMA, exponential smoothing) to multivariate machine-learning forecasting. You leave able to choose the right model for a series and quantify how good the forecast really is.

You work on real time series — decomposing them, fitting ARIMA and smoothing models, then moving to machine-learning and multivariate forecasting and anomaly detection. The capstone forecasts real demand or operations data.

• Analysts and data scientists forecasting demand, sales or operations.
• Finance, supply-chain and operations professionals.

• Analyze and decompose real time-series data.
• Build classical forecasts with ARIMA and exponential smoothing.
• Apply machine learning to univariate and multivariate forecasting.
• Detect anomalies in time-series data.
• Select and validate the right forecasting model.

• Hands-on labs throughout the program (70% project-based).
• A capstone project demonstrating end-to-end skills.

• Build accurate forecasts and detect anomalies in real time-series data.
• Choose the right model for univariate and multivariate problems.

Globally accredited certificate issued by Epsilon AI Learning — USA, with a unique Certificate ID and EPSILON ID. Awarded on 80% attendance, 80% final exam, and a capstone project.

• Trend, seasonality and decomposition
• Stationarity and differencing
• Resampling and feature creation

• AR, MA, ARIMA and SARIMA
• Exponential smoothing
• Model selection and diagnostics

• Supervised framing of time series
• Tree- and boosting-based models
• Backtesting and validation

• Multivariate forecasting
• External regressors and drivers
• Anomaly detection in time series

• Error metrics (MAE, RMSE, MAPE)
• Forecast intervals and uncertainty
• Choosing between models

• Forecasting real demand or operations
• Detecting anomalies in the series
• Communicating the results