CEGM1000 MUDE: Week 2.4, Friday, Dec 6, 2024.
You can access this assignment with the following link: classroom.github.com/a/y8V5r7AZ.
The focus of this assignment is on time series analysis.
Your primary objective is to complete all tasks in the notebook Analysis.ipynb. Unlike other weeks, for this GA it is not required to put your answers in a Report.md file (yay!).
Make sure you use the space provided in the notebooks to complete the tasks, and that the output is included in the notebook when you commit and push it to GitHub.
README.md with instructionsAnalysis.ipynbtemperature.csv: data fileYou can complete this assignment with the same environment you used for WS 2.4 earlier this week (the only special package relative to other weeks is statsmodels).
Grading and submission is identical to previous weeks. Note, however that the Tasks in the notebook that are very similar to WS 2.4 are not graded.
As mentioned above: there is no Report.md, only a notebook!
This assignment overlaps with WS 2.4 in terms of identifying time series components and removing them iteratively. Unlike WS 2.4, however, we will consider an offset (Part 3). It may be possible for some group members to start reading and studying this part while the others are implementing the first part of the notebook.
The most challenging part of the code will be the implementation of the offset detection algorithm in Part 3. Also different from Wednesday will be the implementation of AR(2), both to evaluate the residuals in Part 4 and to create and evaluate the functional model in Part 5.
End of file.
By MUDE Team © 2024 TU Delft. CC BY 4.0. doi: 10.5281/zenodo.16782515.