Term project 1 of Maching Learning 2025-1

In this project, you will train regression models to predict KOSPI.

Datasets

The training dataset has KOSPI from 2019 to 2022. The test dataset has KOSPI in 2023. See load_data.ipynb to load the datasets.

Part 1. Predict KOSPI of the next day.

Train regression models to predict the next day's close using Open, Low, High, Close, Volume of previous days as predictors using only the training dataset. Cross-validate to select the best model. Evaluate the accuracy of your model using the test dataset.

Part 2. Use more predictors to improve your model.

Extend the regression model by adding some extra predictors of your choice. You can use any statistics publicly available.

Report template

• Report must be a pdf file with filename YOUR_STUDENT_ID.pdf
• Summary (maximum 250 words): provide a brief overview, the problem statement, methodology, findings, and conclusions
• Introduction: provide background, define the problem, and state your objectives
• Methods: provide technical details (preproprocessing data, models used, details of experiments, how to analyze the results). Include the link to your GitHub repository. If you want to keep yor repository private, invite ssuai as a collaborator.
• Results: present findings (Use graphs or tables!)
• Discussion: interpret your findings and discuss implications
• Conclusion: summarize main findings, provide the main message, discuss future directions
• References: list of all the sources cited in the report