Part 2: Statistikklubben LAB - Sales Prediction with XGBoost

NOTE 1: This is part 2 of the lab. If you haven't attended the first lab, please
read the "New to the Lab" section below on how to prepare.
NOTE 2: Spots are limited. If something comes up, please give up your spot
so others can attend ❤️

LAB:
In this hands-on lab, we will work with the same basic setup, built around
two core components:
Model: XGBoost
Task: Predicting sales outcomes

Level
Basic knowledge of Python and some familiarity with machine learning or
statistical modeling is recommended.

XGBoost
XGBoost is a natural choice for this kind of project. It's powerful, flexible, well-documented and widely used in real-world forecasting tasks, yet still
accessible enough that you don't need deep domain expertise to get
started. The sales-prediction problem itself strikes a nice balance: realistic,
but not so specialized that it requires industry knowledge. We'll go through
the notebook and experiment with different settings to understand how
they affect the model. The code covers:

Data preprocessing Exploratory analysis
Feature engineering Handling missing values and outliers
Fine-tuning Evaluation

Comparing results across different approaches can help us see how different settings affect model performance. The goal is not just to build
a model, but to learn from each other, compare methodologies, and deepen
our understanding of predictive modeling as a whole.

New to the Lab?
Once you've signed up, I'll send you code + data + a list of core concepts.

Make sure you can run the notebook without errors before the lab. If
you get errors, check that all required tools, libraries/packages are
installed. Try to get an overview of the concepts and terms (read, google, or ask an AI). You don't need to have a solid grasp of the concepts and
terms but be aware that we won't be going through these during the
lab. Full focus on the code and optimizing the model.

• The notebook is in Python, but an R draft is also avai