Velora

Velora is an AI-powered stochastic market forecasting platform built with React + Vite. It models stock terminal price distributions with Monte Carlo simulation (up to 1,000 paths), Geometric Brownian Motion, and a 3-state Markov chain, then dynamically adjusts risk assumptions using Groq LLaMA 3.3 sentiment analysis on live headlines.

Core capabilities

• Live OHLCV ingestion from Twelve Data and company headline context from Finnhub
• Extra live data pack: quote change %, day range, market cap, and P/E (TTM) from Finnhub
• Monte Carlo engine with regime-switching volatility and jump-shock stress events
• Groq LLaMA 3.3-70B contextual sentiment synthesis to influence drift/vol assumptions
• Native HTML5 Canvas fan chart and histogram rendering (no charting library dependency)
• Risk analytics: expected terminal price, upside probability, VaR(95%), CVaR(95%)
• Regime occupancy heatmap and Markov transition matrix diagnostics
• Scenario presets (Base, Risk-On, Risk-Off, Black Swan) that jointly reweight drift/vol/shock assumptions
• Technical indicator panel: SMA20/50, RSI(14), realized volatility, max drawdown, volume regime
• Multi-horizon probability cone table across forecast checkpoints

Setup

1. Install dependencies:

   npm install

2. Configure environment:

   cp .env.example .env

3. Add API keys in .env:

   • VITE_TWELVE_DATA_API_KEY
   • VITE_FINNHUB_API_KEY
   • VITE_GROQ_API_KEY

4. Run:

   npm run dev

If any key is missing, Velora automatically runs in demo mode with synthetic data and fallback sentiment.