[Stock Market Analytics Zoomcamp 2025] Module 3 "Time Series Modeling"

Links:
📂 Course GitHub: https://github.com/DataTalksClub/stock-markets-analytics-zoomcamp/blob/main/README.md
📝 Registration Form: https://forms.gle/qP19FoBSnzmZsJYt6
🌐 PythonInvest Website: https://pythoninvest.com/
☕ Support PythonInvest: https://buymeacoffee.com/pythoninvest OR https://github.com/sponsors/realmistic

Module 3: Time Series Modeling & Dataset Preparation
In this session, we move from data preparation to building and interpreting time series prediction models. You'll learn how to construct a unified dataset, apply tree-based models, and explain your forecasts in a structured and meaningful way.

🔍 Topics Covered:

Data Preparation & Feature Engineering
* Defining key variable sets
* Creating dummy variables for categorical features
* Temporal train/test split
* Correlation analysis for feature selection

Time Series Modeling
* Simple “hand-rule” based forecasting
* ARIMA for statistical baseline predictions
* Binary Decision Tree as the main modeling technique
* Understanding and applying model explainability techniques

⏱ Timestamps:
0:00 - Introduction & Homework Comments
Opening remarks, session agenda, overview of homework results, and participant feedback.

16:33 - Data Preparation & Feature Engineering
Unifying datasets, defining variable sets, creating dummy variables for categorical data, and initial preprocessing steps.

28:40 - Correlation Analysis & Feature Selection
Performing correlation analysis for all features, discussing relevance, data splitting for train/test/validation, and properly structuring time series data.

46:44 - Time Series Modeling & Hand Rules
Introduction to hand-rule based forecasting, overview of ARIMA/statistical baselines, implementing and interpreting manual heuristics.

1:13:10 - Tree-Based Models & Explainability
Building, evaluating, and explaining binary decision tree models; discussion of feature importances, overfitting, and practical ML model explainability.

1:25:35 - Recap & Next Steps
Summary of key learnings, preview for the next module, and closing comments.
#PythonInvest #Finance #TimeSeries #Forecasting #DataScience #Investing #MachineLearning #ARIMA #DecisionTree #ModelExplainability #StockMarket #GoogleColab

Видео [Stock Market Analytics Zoomcamp 2025] Module 3 "Time Series Modeling" канала PythonInvest