@dwlf/indicators

Zero-dependency technical indicators for Node.js, browsers, and workers.

Pure math — candles in, values out. No market data, no side effects, no runtime dependencies.

Install

npm install @dwlf/indicators
# or
pnpm add @dwlf/indicators

Quick Start

import { DSS, Bollinger, EMA, SupportResistance } from '@dwlf/indicators';
import type { Candle } from '@dwlf/indicators';

const candles: Candle[] = [
  { t: 1700000000, o: 100, h: 105, l: 98, c: 103, v: 1000 },
  // ...
];

// Compute indicators — all functions take (candles, optionalParams)
const ema8 = EMA.computeEMA(candles, { length: 8 });
const dss = DSS.computeDSS(candles, { length: 10, smooth1: 9, signal: 5 });
const bb = Bollinger.computeBollingerBands(candles, { length: 20, standardDeviation: 2 });

// Detect events
const dssEvents = DSS.detectEvents(candles);
const emaEvents = EMA.detectEvents(candles, { length: 8 });

Indicators

| Module | Description | |--------|-------------| | DSS | Double Smoothed Stochastic (Bressert) | | DSSState | Streaming DSS state tracking | | Bollinger | Bollinger Bands (upper, middle, lower, %B, bandwidth) | | EMA | Exponential Moving Average with event detection | | SMA | Simple Moving Average with event detection | | EMACloud | EMA cloud/ribbon (alignment, cloud hit) | | Swing | Swing high/low detection | | SwingBreak | Swing break detection | | SwingSweep | Liquidity sweep detection | | Fib | Fibonacci retracement levels | | SupportResistance | Support/resistance level calculation | | SupportResistanceTracker | S/R tracking over time | | Trendline | Trendline detection | | BreachDetection | Trendline breach detection |

Math Utilities

Low-level building blocks, also exported:

import { ema, sma, rollingHighest, rollingLowest, standardDeviation } from '@dwlf/indicators';

API Pattern

Every indicator follows the same pattern:

// Compute raw values
const result = Module.computeX(candles, params?);

// Detect events (crossovers, threshold breaches, etc.)
const events = Module.detectEvents(candles, params?);

// Get event metadata (what events this indicator can emit)
const definitions = Module.getEventDefinitions();

BYOC: Bring Your Own Candles

This library is intentionally data-source agnostic. Provide candles in the standard OHLCV format:

type Candle = {
  t: number;  // timestamp (unix seconds or ms)
  o: number;  // open
  h: number;  // high
  l: number;  // low
  c: number;  // close
  v?: number; // volume (optional)
};

Used By

This is the same indicator math that powers DWLF — a market intelligence platform for traders. The event detection, strategy engine, and cycle analysis that sit on top of these indicators are available via the DWLF platform.

License

MIT