It’s code breakfast time! This time we meet online, but with the positive vibe kept and some breakfast recipes shared! Here is what we will talk and code about.
From inventory to website visitors, resource planning to financial data, time-series data is all around us. Knowing what comes next is key to success in this dynamically changing world. And for that, we need reliable forecasting models. While complex & deep models may be good at forecasting, they typically give us little insight into the underlying patterns in our data. Such insights, however, may be a key to not only forecasting the future but shaping it.
In this code breakfast, we will from scratch build relatively simple models that can give us such insights. We will find out why understanding seasonality is important and what data can actually tell us about it. Among other things we will learn about:
- what a time series consists of; how to decompose it and why
- naïve approaches to detecting seasonality and related dangers
- using our insights to turn simple models into powerful tools
- building blocks for seasonality from dummies to Fourier series
- fine-tuning, evaluating and interpreting models
- dealing with overfitting and other hidden challenges
- and much more!