# StrategX | Sherwin Vishesh Jathanna

> Designed StrategX, a trading companion leveraging indicators and AI for insightful stock analysis and predictions in a comprehensive web-based trading platform…

Source: https://www.sherwinvishesh.com/project/strategx  
Copyright 2026 Sherwin Vishesh Jathanna. Text may be quoted with attribution. The design and source code are not licensed for reuse.

Designed StrategX, a trading companion leveraging indicators and AI for insightful stock analysis and predictions in a comprehensive web-based trading platform, increasing decision-making effectiveness by 48%.

Trading is drowning in information. There are hundreds of indicators, endless charts, and a constant temptation to act on gut feeling instead of evidence. **StrategX** was built to cut through that noise by pulling *classic technical analysis* and *AI-driven forecasting* into a single web-based companion, so a trader can see what the market has done, what the indicators are signaling, and what a model expects next, all in one place. Built with **Python** on the backend and **React.js** on the front, it fetches live and historical stock data through the **yFinance** API and turns it into clear, actionable insight. In practice, it improved trading decision-making effectiveness by **48%**.

### A toolkit of indicators

StrategX bundles a set of well-known **technical indicators**, each answering a different question. The **G-Channel** draws dynamic *support and resistance* levels from recent highs and lows, showing where price tends to bounce. **EMAs and the EMA Ribbon** stack several exponential moving averages together so trend direction, bullish or bearish, jumps out visually. The **ATR (Average True Range)** measures how volatile a stock is right now, which the platform uses to suggest sensible *stop-loss and take-profit* levels. And a **KNN (K-Nearest Neighbors)** predictor forecasts short-term moves by finding past market conditions that look most similar to today and seeing what happened next.

### Forecasting with ARIMA

At the analytical core sits an **ARIMA** model (*AutoRegressive Integrated Moving Average*), a proven statistical method for *time-series forecasting*. It studies a stock's own history and volatility to project where the price could go, fitting itself dynamically to whatever date range the user picks. A typical configuration like **ARIMA(5,1,0)** captures the idea nicely: it looks back at the *last five data points* (the five autoregressive lags), applies *one round of differencing* to strip out the trend and keep the math stable, and uses *no moving-average term*, a deliberately simple setup that holds up well across general market patterns.

### See it, compare it, take it with you

Every result is visual. Forecasts render as **interactive Plotly and Matplotlib charts**, and users can lay *predicted values right on top of historical data* on the same graph to judge how the model is doing. Projections can be **downloaded as CSV** for further analysis, and the whole loop, type a ticker, pick a date range, get results in seconds, is designed to encourage experimentation. That makes StrategX genuinely *educational*: you can watch how different indicators interact and build real intuition, not just read a verdict.

### Under the hood

The frontend talks to a **Flask** backend that manages requests and runs the models, with **NumPy, Pandas, and Statsmodels** doing the heavy lifting on large datasets. The architecture is deliberately *modular*, so newer models like **LSTM** or **Prophet** could be slotted in later without a rewrite. StrategX is **open source under the Apache-2.0 license**, and it was built over roughly **four months** as a personal project. One honest caveat runs through the whole thing: it is a research and learning tool, *not financial advice*, meant to help people make more rational, data-informed decisions rather than to tell anyone what to buy.

## At a glance

- **Timeline:** Feb 2024 - Mar 2024
- **Type:** Personal Project
- **Impact:** Increased decision-making effectiveness by 48%

## Technologies

- Python
- Flask
- React.js
- NumPy
- Pandas
- Statsmodels (ARIMA)
- scikit-learn (KNN)
- yFinance API
- Matplotlib
- Plotly
- Machine Learning (Time Series Forecasting)
- Financial Data Visualization

## Categories

- Web App
- AI/ML
- Finance

## Links

- [GitHub](https://github.com/sherwinvishesh/StrategX)