VN1 Forecasting - Accuracy Challenge Phase 1
Share:
Flieber, Syrup Tech, and SupChains Launch an AI-Driven Supply Chain Forecasting Competition
This notebook demonstrates univariate time series modeling using the fable package in R. It includes data preparation, exploratory analysis, and the development of multiple forecasting models (sNaive, ARIMA, ETS, Mean, and Drift) with parallel processing, followed by forecast accuracy evaluation.
This notebook provides a comprehensive guide to univariate time series modeling using the fable package, incorporating parallel processing for improved efficiency. The workflow starts with the preparation of sales data, including tidying and transforming it into a tsibble format suitable for time series analysis. Exploratory analysis is conducted to check for missing values and temporal gaps in the data.
The notebook then splits the time series data into training and test sets and builds various forecasting models, such as sNaive, ARIMA, ETS, Mean, and Drift models. These models are fitted using parallel processing with foreach and doParallel, enabling faster computation across multiple cores. Finally, forecast accuracy is evaluated using metrics like Mean Error (ME), Root Mean Squared Error (RMSE), and Mean Absolute Error (MAE), providing a summary of each model's performance.
This notebook is ideal for those looking to implement time series forecasting with scalable performance in R using the fable framework.
Read more: https://otexts.com/fpp3/
Keep up to date by participating in our global community of data scientists and AI enthusiasts. We discuss the latest developments in data science competitions, new techniques for solving complex challenges, AI and machine learning models, and much more!
Nicolas Vandeput
Posted 3 months ago