Saturday Morning Data Science Projects

Collection of quick & dirty data science projects aimed to investigate topics that I find interesting.

Title inspired from Andriew Ng's quote: "Fun" is what you do on Saturday morning.

📊 Projects on nbViewer

01 - Will I ever retire in Italy?

🛠️ istatapi, open-meteo

All main results are summarized in this website.
The following links bring to the visualization of the single notebooks:

1. Investigating Italian population
2. What's the killer of 2003?
3. Modelling the future population
4. Estimating the sustainable retirement age
5. Split population growth rate in mortality and migration balance

DEPRECADED: in June 2024 I updated the project by spliting the mortality rate and migration balance, reorganizing the notebooks, and hosting the updated project in Demographic Model Page.

02 - Can I time the market (S&P 500)?

🛠️ yfinance, cot_reports

1. Visualizing Ashwat Damodaran's analysis
2. Best US portfolios with ca. 100y of backtesting
3. McQuarrie's thesis on similar returns for stocks and long term bonds
4. Comparing COT report data with S&P500 returns
5. Do I beat the market if I enter when it is -XXX% and axit when it is +YYY%?

03 - Which is the most efficient healty diet?

🛠️ esselunga-scrape, Project Nutrition

1. Scraping Esselunga's database in search of the foods with highest protein/euro ratio
2. Visualizing Jackson-Pollock equations for BMI estimation

📌 Rules

1. State the AIM of the analysis at the beginning of the project
2. End the notebook with a Conclusions section, summarizing the findings, and a Follow-up section, listing possible next steps to take
3. Make intense use of GitHub Copilot
4. Explore creative uses of plotly for interactive visualizations
5. Don't spend more than 2 hours on a project - this is a Saturday morning activity, not a full-blown project that will take weeks to complete
6. The previous point was totally sugested by Copilot... I'm not sure if I agree with it, but it's a fair reminder
7. Don't be too strict in using statisticly sound methods (for sake of time), but be open to critics in case the analisis was too handwaving
8. Come back from time to time, to revise the narrative of the analysis to make it clearer and clearer
9. Be totally open to comments, suggestions, and collaborations: for any of them, open an Issue or a Pull Request
10. Don't have more than 10 rules