README.md

---
title: Multi-Indicator Z-Score Scorer
emoji: 📊
colorFrom: indigo
colorTo: purple
sdk: gradio
sdk_version: 4.44.1
python_version: 3.11
app_file: app.py
pinned: false
license: mit
---

# Multi-Indicator Z-Score Scorer

Pure Pine Script port — **no ML, no entity tracking**. Technical analysis only.

## How it works

1. Compute 4 indicators on close prices:
   - **RSI** (default 14)
   - **MACD histogram** (12/26/9)
   - **Stochastic %K** (14/3)
   - **Trend** (close − EMA 50)

2. Normalize each to a **z-score** over a rolling lookback window (default 100)
3. Weighted sum = **composite z-score**
4. Signals:
   - **Buy** when composite crosses above `buy_level` (default 1.5)
   - **Sell** when composite crosses below `sell_level` (default −1.5)

## Input formats

**CSV upload** — needs a `close` column, optional `high` and `low`

**Manual prices** — comma or newline separated numbers (min 20)

## Source

Python port of `multi_indicator_zscore.pine`. All indicator math matches Pine's built-ins:
`ta.rsi`, `ta.macd`, `ta.stoch`, `ta.ema`, `ta.sma`, `ta.stdev`, `ta.crossover`, `ta.crossunder`.