A beginner-friendly, data-driven guide to investing on the NASDAQ.
Learn key financial concepts, explore real market data, and simulate your own investment strategies.
🌐 Deep Dive in the Finance World Now
https://com-480-data-visualization.github.io/StocksWise/
| Name | SCIPER |
|---|---|
| Gustave Lapierre | 326066 |
| Youssef Seddik | 346488 |
| Maxime Garambois | 346526 |
Milestone 1 - Milestone 2 - Milestone 3
Deadline: Friday, April 3rd – 17:00
Main dataset
https://www.kaggle.com/datasets/jacksoncrow/stock-market-dataset
Back in the early 1980s, a vast financial record began taking shape. Sourced through Yahoo Finance and compiled by Oleh Onyshchak (Kaggle user jacksoncrow), the NASDAQ Stock Market Dataset tracks daily price movements for all NASDAQ-listed securities. Each ticker comes with structured CSV entries: Date, Open, High, Low, Close, Adjusted Close, and Volume. Coverage runs up to April 2020. We plan to extend it to 2025 by forking and re-executing the original collection script. The dataset is very complete, and all rows are relevant. This update will include critical market phases the 2022 sell-off triggered by rate hikes, and the surge in 2023,2024 fueled by AI enthusiasm. It’s a necessary step. Markets evolve. So should the data.
Extended dataset
| Source | What it adds | How |
|---|---|---|
| FinanceDatabase | Sector & industry tags | CSV join on ticker |
| yfinance API | Company name, description | Python queries |
| FRED | Fed interest rates | Free API |
| VIX | Market fear index | Yahoo Finance |
Data quality
- Consistent CSV format (simple for parse)
- Missing values handled with forward-fill or row drop depending on gap size
- Adj Close mandatory for long-term comparisons (e.g. Apple's 2020 4-for-1 split distorts raw prices).
Existing platforms (Bloomberg, Yahoo Finance, Trade Republic) are built for experienced investors dense, jargon-heavy, and intimidating for beginners. Financial literacy is rarely taught in school, yet young adults are increasingly expected to make consequential money decisions: student loans, savings, first investments. The gap between "I have some money" and "I know what to do with it" has never been wider.
Target audience: students aged 18–25 with no financial background and more.
Core questions:
- What investment strategies exist, and what risk/reward does each carry?
- How have companies or sectors evolved on the NASDAQ over time?
- What would €1,000 invested in Apple in 2010 look like today?
- How do ETFs compare to individual stocks?
- What tools help users go further once they know the basics?
StocksWise guides beginners through investing via data storytelling. Rather than overwhelming users with raw numbers, it builds intuition progressively turning price charts into stories, and abstract percentages into personal outcomes.
Core narrative: "You have €1,000 what do you do?" The platform grows with the user through a two-mode architecture.
| Beginner Mode | Advanced Mode |
|---|---|
| Guided step-by-step narrative | Advanced Concepts |
| Strategy simulator | Technical indicators (MA, RSI, Sharpe) |
| Plain-language explanations | Advanced risk / return scatter plot |
| Risk quiz → personalized recommendation | ML clustering: stock risk profiles |
| Metric | Value |
|---|---|
| Stock tickers | 6,800 CSV files |
| ETF tickers | 2,100 CSV files |
| Longest history | 1980–2025 |
| Avg rows / ticker | 3,500 |
| Total rows | 30 million |
Key observations:
- Prices range from 1 to 500 dollars —> normalization mandatory
- Volume spikes mark major crises: dot-com 2000, 2008, COVID 2020, rate hikes 2022.
- Tech (Apple, NVIDIA, Microsoft, Amazon, Meta, Alphabet) dominates by market cap and return.
- ETFs like QQQ are far smoother than individual growth stocks (Tesla, NVIDIA)
- NVIDIA surged +600% in 2023–2024 on AI hype, a compelling story for young audiences.
Tech giants vastly outperform the QQQ index. NVIDIA and Tesla show explosive but volatile growth, while QQQ offers a smoother ride, a key insight for beginners weighing risk vs. reward.
Most tickers have 3,000–10,000 trading days of history. ETFs cluster tightly around the same range, while stocks show a wider spread, some IPO'd recently, others date back to the 1980s.
Individual growth stocks (NVIDIA, Tesla) exhibit 2–4x the volatility of the QQQ ETF, confirming that diversified ETFs offer significantly smoother returns, a core lesson for our target audience.
Preprocessing plan: sector tagging (FinanceDatabase), price normalization, rolling metrics (MA 30d/200d, Sharpe, volatility), monthly aggregation for long-range views. Raw prices are adjusted for splits and dividends using Adj Close to ensure longitudinal consistency. Tickers with fewer than 252 trading days (one full year) are excluded to guarantee meaningful trend analysis. For volume data, extreme outliers are capped at the 99th percentile to prevent crisis spikes from distorting visual scales. A unified schema is enforced across all CSV files before ingestion, mapping each ticker to its sector, industry, and asset type (stock vs. ETF). This structured pipeline ensures that every visualization from simple line charts to ML-driven recommendations is built on clean, comparable, and pedagogically meaningful data.
| Tool | What they do | What's missing |
|---|---|---|
| Yahoo Finance / Bloomberg | Comprehensive but expert-only | No education, no personalization |
| Trade Republic | Simple trading app | No visual explanation of concepts |
| NYT The Upshot | Narrative-driven viz reference | Not finance-focused |
| Morningstar | Risk/return plots | Not for beginners |
Why we are original:
- Only platform combining education + exploration + simulation + ML personalization for beginners.
- Two mode architecture is new: the app grows with the user.
- The "€1,000" narrative makes abstract concepts personal and memorable.
See PDF "Milestone 2"