TrendCraft

A TypeScript library for technical analysis of financial data. Calculate indicators, detect signals, and analyze market trends.

日本語版 README

Features

Indicators (130+)

• Moving Averages: SMA, EMA, WMA, VWMA, KAMA, T3, HMA (Hull), McGinley Dynamic, EMA Ribbon, DEMA, TEMA, ZLEMA, FRAMA, ALMA
• Trend: Ichimoku Cloud, Supertrend, Parabolic SAR, Vortex, Schaff Trend Cycle, Linear Regression
• Momentum: RSI, MACD, Stochastics (Fast/Slow), DMI/ADX, ADXR, Stoch RSI, CCI, Williams %R, ROC, TRIX, Aroon, DPO, Hurst Exponent, Connors RSI, IMI, Ultimate Oscillator, Awesome Oscillator, Mass Index, KST, Coppock Curve, TSI, PPO, CMO, Balance of Power, QStick
• Volatility: Bollinger Bands, ATR, Donchian Channel, Keltner Channel, Chandelier Exit, Volatility Regime, Choppiness Index, Ulcer Index, Historical Volatility, Garman-Klass, Standard Deviation
• Volume: OBV, MFI, VWAP (with Bands), Volume MA, CMF, Volume Anomaly, Volume Profile, Volume Trend, ADL, Anchored VWAP, Elder Force Index, Ease of Movement, Klinger, TWAP, Weis Wave, Market Profile, PVT, NVI
• Price: Highest/Lowest, Returns, Pivot Points, Swing Points, Fractals, Zigzag, Fibonacci Retracement/Extension, Heikin-Ashi, Opening Range Breakout, Gap Analysis, Median Price, Typical Price, Weighted Close
• Smart Money Concepts (SMC): Order Block, Liquidity Sweep
• Filter: Super Smoother, Roofing Filter (Ehlers)
• Relative Strength: Benchmark RS, RS Rating, RS Ranking

Signal Detection

• Cross Detection: Golden Cross, Dead Cross, custom crossovers
• Fake Signal Detection: Validate cross signals with volume/trend confirmation
• Divergence: OBV, RSI, MACD divergence detection
• Squeeze: Bollinger Bands squeeze detection
• Range-Bound: Detect sideways markets and potential breakouts
• Chart Patterns: Double Top/Bottom, Head & Shoulders, Cup with Handle, Triangle (symmetrical/ascending/descending), Wedge (rising/falling), Channel (ascending/descending/horizontal), Flag/Pennant
• Pattern Filter: ATR ratio, trend context, and volume confirmation filters

Backtesting

• Simple strategy backtesting with preset conditions
• Stop loss, take profit, trailing stop support
• Commission and slippage simulation
• Performance metrics (Sharpe ratio, max drawdown, win rate)
• Multi-timeframe (MTF) conditions (weekly/monthly RSI, SMA, trend)
• Advanced volume conditions (anomaly detection, volume profile)

Signal Scoring

• Combine multiple signals with weighted importance
• Pre-built presets: momentum, meanReversion, trendFollowing, balanced
• Fluent API (ScoreBuilder) for custom scoring strategies
• Backtest conditions based on composite scores

Position Sizing

• Risk-Based: Calculate size from risk amount and stop distance
• ATR-Based: Dynamic stop distance using ATR
• Kelly Criterion: Optimal sizing based on win rate and payoff ratio
• Fixed Fractional: Simple percentage-based allocation

ATR Risk Management

• Chandelier Exit trailing stops
• Dynamic ATR-based stop and take-profit levels
• Integrated with backtesting engine

Strategy Optimization

• Grid Search: Parameter tuning with constraints
• Walk-Forward Analysis: Out-of-sample validation
• Combination Search: Find optimal entry/exit condition combinations

Scaled Entry

• Split entry strategies (equal, pyramid, reverse-pyramid)
• Signal-based or price-based intervals
• Partial take-profit support

Volatility Regime Detection

• Classify market volatility (low, normal, high, extreme)
• Filter trades by volatility environment
• ATR% filtering for stock screening

CLI Tools

• Screening CLI: Screen multiple stocks against entry/exit conditions
• Backtest CLI: Run backtests on single files or compare across multiple files
• Output formats: table, JSON, CSV

Utilities

• Data normalization (various date formats to timestamps)
• Timeframe resampling (daily to weekly/monthly)
• Fluent API for chaining operations

Signal Explainability

• Trace why signals fire with structured condition traces
• Capture indicator values at evaluation time
• Human-readable narratives in English and Japanese
• Works with any combination of preset, combined, and MTF conditions

Composable Indicator Algebra

• pipe() / compose() / through() API for chaining indicators
• Apply any indicator to another indicator's output: ema(rsi(candles, 14), 9)
• Extract fields from complex indicators (MACD histogram, Bollinger %B)
• Combine and transform series with mapValues(), combineSeries()

Alpha Decay Monitor

• Track strategy predictive power degradation over time
• Rolling Information Coefficient (IC) and hit rate
• CUSUM structural break detection
• Four-level assessment: healthy, warning, degraded, critical

Adaptive Indicators

• Adaptive RSI: Period adjusts based on ATR volatility percentile
• Adaptive Bollinger Bands: StdDev multiplier adjusts based on kurtosis
• Adaptive MA: Smoothing adapts via Efficiency Ratio (trending vs choppy)
• Adaptive Stochastics: Period adjusts based on ADX trend strength

Strategy Robustness Score

• Composite A+ to F grade from multiple dimensions
• Monte Carlo survival rate, trade consistency, drawdown resilience
• Full analysis with parameter sensitivity and walk-forward efficiency
• Actionable recommendations for improvement

Pairs Trading / Cointegration

• Engle-Granger cointegration test (ADF)
• Spread calculation with z-score (full sample or rolling window)
• Mean reversion half-life estimation (AR(1) model)
• Hurst exponent via rescaled range analysis
• Automated pairs trading signal generation

Cross-Asset Correlation

• Rolling Pearson and Spearman correlation
• Correlation regime detection (5 regimes with duration tracking)
• Lead-lag analysis via cross-correlation at multiple lags
• Intermarket divergence detection with z-score significance

TA-Lib Cross-Validation

36 indicators are cross-validated against TA-Lib (Python ta-lib 0.6.8) using 200-bar synthetic OHLCV data covering 4 market phases (uptrend → high volatility → range → downtrend). Test code and fixtures are in the cross-validation/ directory.

| Precision | Indicators | Decimals | |-----------|-----------|----------| | Exact | SMA, Highest, Lowest, Donchian Channel, Median Price, Typical Price, Weighted Close, OBV, ADL, Standard Deviation | 9+ | | High | EMA, WMA, RSI, CCI, Williams %R, ROC, ATR, MFI, DMI/ADX, Keltner Channel, Bollinger Bands, KAMA, Parabolic SAR, DEMA, TEMA, CMO, Aroon, ADXR, Ultimate Oscillator, Linear Regression | 6–8 | | Good | MACD, Stochastics (Fast/Slow), T3, StochRSI, PPO | 3–4 |

Good tier note: Minor differences are due to documented implementation differences (e.g., MACD EMA seeding, T3 cascaded EMA warmup). These converge over longer series.

Installation

npm install trendcraft

Quick Start

import { sma, rsi, macd, bollingerBands } from 'trendcraft';

const candles = [
  { time: 1700000000000, open: 100, high: 105, low: 99, close: 104, volume: 1000 },
  // ... more candles (OHLCV format)
];

// Calculate indicators — each returns Series<T> = { time, value }[]
const sma20 = sma(candles, { period: 20 });
const rsi14 = rsi(candles, { period: 14 });
const bb    = bollingerBands(candles, { period: 20, stdDev: 2 });

// Backtest a strategy
import { TrendCraft, goldenCrossCondition, deadCrossCondition, and, rsiBelow } from 'trendcraft';

const result = TrendCraft.from(candles)
  .strategy()
  .entry(and(goldenCrossCondition(), rsiBelow(50)))
  .exit(deadCrossCondition())
  .backtest({ capital: 1_000_000, stopLoss: 5, takeProfit: 15 });

console.log(`Return: ${result.totalReturnPercent.toFixed(2)}%  Sharpe: ${result.sharpeRatio.toFixed(3)}`);

Runnable examples — see examples/quick-start/ for 5 self-contained scripts covering indicators, backtesting, optimization, screening, and streaming.

Usage Examples

Moving Averages

import { sma, ema } from 'trendcraft';

// Simple Moving Average
const sma20 = sma(candles, { period: 20 });
// Returns: [{ time: number, value: number | null }, ...]

// Exponential Moving Average
const ema12 = ema(candles, { period: 12 });

RSI

import { rsi } from 'trendcraft';

const rsi14 = rsi(candles, { period: 14 });

// Check oversold/overbought
rsi14.forEach(({ time, value }) => {
  if (value !== null) {
    if (value < 30) console.log(`${time}: Oversold`);
    if (value > 70) console.log(`${time}: Overbought`);
  }
});

MACD

import { macd } from 'trendcraft';

const macdData = macd(candles, { fast: 12, slow: 26, signal: 9 });

macdData.forEach(({ time, value }) => {
  const { macd: macdLine, signal, histogram } = value;
  // macdLine: MACD line value
  // signal: Signal line value
  // histogram: MACD - Signal
});

Bollinger Bands

import { bollingerBands } from 'trendcraft';

const bb = bollingerBands(candles, { period: 20, stdDev: 2 });

bb.forEach(({ time, value }) => {
  const { upper, middle, lower, percentB, bandwidth } = value;
  // upper: Upper band
  // middle: Middle band (SMA)
  // lower: Lower band
  // percentB: %B indicator (0-1 scale)
  // bandwidth: Band width
});

Signal Detection

import { goldenCross, deadCross, rsiDivergence, bollingerSqueeze } from 'trendcraft';

// Golden Cross / Dead Cross
const gc = goldenCross(candles, { short: 5, long: 25 });
const dc = deadCross(candles, { short: 5, long: 25 });

// RSI Divergence
const divergences = rsiDivergence(candles);
divergences.forEach(signal => {
  console.log(`${signal.type} divergence at ${signal.time}`);
  // type: 'bullish' or 'bearish'
});

// Bollinger Squeeze
const squeezes = bollingerSqueeze(candles, { threshold: 10 });
// Detects low volatility periods (potential breakout setup)

// Range-Bound Detection
import { rangeBound } from 'trendcraft';

const rb = rangeBound(candles);
const latest = rb[rb.length - 1].value;

if (latest.state === 'RANGE_CONFIRMED') {
  console.log(`Range: ${latest.rangeLow} - ${latest.rangeHigh}`);
}
if (latest.state === 'BREAKOUT_RISK_UP') {
  console.log('Watch for upside breakout!');
}

Timeframe Resampling

import { resample } from 'trendcraft';

// Convert daily candles to weekly
const weeklyCandles = resample(dailyCandles, 'weekly');

// Convert to monthly
const monthlyCandles = resample(dailyCandles, 'monthly');

Fluent API

import { TrendCraft } from 'trendcraft';

const result = TrendCraft.from(candles)
  .sma(20)
  .ema(12)
  .rsi(14)
  .compute();

// Access results
console.log(result.sma);
console.log(result.ema);
console.log(result.rsi);

Backtesting

import { TrendCraft, goldenCross, deadCross, and, rsiBelow } from 'trendcraft';

// Simple backtest with preset conditions
const result = TrendCraft.from(candles)
  .strategy()
    .entry(goldenCross())        // Enter on golden cross
    .exit(deadCross())           // Exit on dead cross
  .backtest({ capital: 1000000 });

console.log(result.totalReturnPercent);  // Total return %
console.log(result.winRate);             // Win rate %
console.log(result.maxDrawdown);         // Max drawdown %
console.log(result.sharpeRatio);         // Sharpe ratio

// Combine conditions with AND/OR
const advancedResult = TrendCraft.from(candles)
  .strategy()
    .entry(and(goldenCross(), rsiBelow(30)))  // GC + RSI oversold
    .exit(deadCross())
  .backtest({
    capital: 1000000,
    stopLoss: 5,       // 5% stop loss
    takeProfit: 15,    // 15% take profit
    commission: 0,
    commissionRate: 0.1,  // 0.1% commission
  });

Volume Analysis

import { volumeAnomaly, volumeProfile, volumeTrend } from 'trendcraft';

// Detect unusual volume spikes
const anomalies = volumeAnomaly(candles, { period: 20, highThreshold: 2.0 });
anomalies.forEach(({ time, value }) => {
  if (value.isAnomaly) {
    console.log(`${time}: ${value.level} volume (${value.ratio.toFixed(1)}x avg)`);
  }
});

// Volume Profile with POC, VAH, VAL
const profile = volumeProfile(candles, { period: 20 });
console.log(`POC: ${profile.poc}`);      // Point of Control (highest volume price)
console.log(`VAH: ${profile.vah}`);      // Value Area High
console.log(`VAL: ${profile.val}`);      // Value Area Low

// Volume Trend Confirmation
const trends = volumeTrend(candles);
trends.forEach(({ time, value }) => {
  if (value.isConfirmed) {
    console.log(`${time}: Trend confirmed (${value.confidence}%)`);
  }
  if (value.hasDivergence) {
    console.log(`${time}: Volume divergence detected`);
  }
});

Multi-Timeframe (MTF) Conditions

import { weeklyRsiAbove, weeklyPriceAboveSma, and, goldenCrossCondition } from 'trendcraft';

// Backtest with weekly RSI filter
const result = TrendCraft.from(dailyCandles)
  .withMtf(['weekly'])  // Enable MTF with weekly data
  .strategy()
    .entry(and(
      weeklyRsiAbove(50),        // Weekly RSI > 50 (bullish bias)
      goldenCrossCondition()     // Daily golden cross
    ))
    .exit(deadCrossCondition())
  .backtest({ capital: 1000000 });

Signal Scoring

import { ScoreBuilder, calculateScore, scoreAbove } from 'trendcraft';

// Build custom scoring strategy
const config = ScoreBuilder.create()
  .addPOConfirmation(3.0)      // Perfect Order confirmation (weight: 3.0)
  .addRsiOversold(30, 2.0)     // RSI < 30 (weight: 2.0)
  .addVolumeSpike(1.5, 1.5)    // Volume > 1.5x average (weight: 1.5)
  .addMacdBullish(1.5)         // MACD bullish crossover
  .setThresholds(70, 50, 30)   // strong, moderate, weak
  .build();

// Calculate score at specific index
const result = calculateScore(candles, candles.length - 1, config);
console.log(result.normalizedScore);  // 0-100 score
console.log(result.strength);         // "strong" | "moderate" | "weak" | "none"

// Use in backtest
import { TrendCraft, deadCross } from 'trendcraft';

const backtestResult = TrendCraft.from(candles)
  .strategy()
    .entry(scoreAbove(70, "trendFollowing"))  // Enter when score > 70
    .exit(deadCross())
  .backtest({ capital: 1000000 });

Position Sizing

import { riskBasedSize, atrBasedSize, kellySize, fixedFractionalSize } from 'trendcraft';

// Risk-based: Calculate size from risk amount and stop distance
const riskResult = riskBasedSize({
  accountSize: 100000,
  entryPrice: 50,
  stopLossPrice: 48,
  riskPercent: 1,           // Risk 1% of account
  maxPositionPercent: 25,   // Max 25% position size
});
// { shares: 500, positionValue: 25000, riskAmount: 1000, ... }

// ATR-based: Dynamic stop using ATR
const atrResult = atrBasedSize({
  accountSize: 100000,
  entryPrice: 50,
  atrValue: 2.5,
  atrMultiplier: 2,         // 2x ATR for stop distance
  riskPercent: 1,
});
// { shares: 200, stopPrice: 45, ... }

// Kelly Criterion: Optimal sizing based on historical win rate
const kellyResult = kellySize({
  accountSize: 100000,
  entryPrice: 50,
  winRate: 0.6,
  winLossRatio: 1.5,
  kellyFraction: 0.5,       // Half Kelly (safer)
});

// Fixed Fractional: Simple percentage allocation
const fixedResult = fixedFractionalSize({
  accountSize: 100000,
  entryPrice: 50,
  fractionPercent: 10,      // 10% of account
});

ATR Risk Management

import { chandelierExit, calculateAtrStop, TrendCraft, goldenCross, deadCross } from 'trendcraft';

// Chandelier Exit trailing stop
const chandelier = chandelierExit(candles, { period: 22, multiplier: 3 });
chandelier.forEach(({ time, value }) => {
  console.log(`Long stop: ${value.longStop}, Short stop: ${value.shortStop}`);
});

// Calculate ATR-based stop and take-profit levels
const stops = calculateAtrStop({
  entryPrice: 100,
  atrValue: 2.5,
  stopMultiplier: 2,        // 2x ATR for stop
  takeProfitMultiplier: 3,  // 3x ATR for take-profit
  direction: 'long',
});
// { stopPrice: 95, takeProfitPrice: 107.5, riskRewardRatio: 1.5 }

// Use in backtest with ATR-based stops
const result = TrendCraft.from(candles)
  .strategy()
    .entry(goldenCross())
    .exit(deadCross())
  .backtest({
    capital: 1000000,
    atrRisk: {
      enabled: true,
      period: 14,
      stopMultiplier: 2,
      takeProfitMultiplier: 3,
    },
  });

Signal Explainability

import { explainSignal, rsiBelow, goldenCrossCondition } from 'trendcraft';

const explanation = explainSignal(candles, 50, rsiBelow(30), goldenCrossCondition());

console.log(explanation.fired);       // true/false
console.log(explanation.narrative);   // "Entry signal fired because rsiBelow(30): passed..."
console.log(explanation.contributions);  // [{ name: "rsiBelow(30)", passed: true, indicatorValues: { rsi14: 28.5 } }]

Composable Indicator Algebra

import { pipe, through, extractField, rsi, ema, macd, bollingerBands } from 'trendcraft';

// EMA of RSI (smoothed RSI)
const smoothedRsi = pipe(
  candles,
  c => rsi(c, { period: 14 }),
  through(ema, { period: 9 }),
);

// Bollinger Bands of MACD histogram
const histBands = pipe(
  candles,
  c => macd(c),
  s => extractField(s, "histogram"),
  through(bollingerBands, { period: 20 }),
);

Alpha Decay Monitor

import { analyzeAlphaDecay, createObservationsFromTrades, runBacktest } from 'trendcraft';

const result = runBacktest(candles, entry, exit, options);
const observations = createObservationsFromTrades(result.trades);
const decay = analyzeAlphaDecay(observations);

console.log(decay.assessment.status);    // "healthy" | "warning" | "degraded" | "critical"
console.log(decay.assessment.halfLife);  // Estimated bars until IC halves

Adaptive Indicators

import { adaptiveRsi, adaptiveBollinger } from 'trendcraft';

// RSI that uses shorter period in high volatility
const arsi = adaptiveRsi(candles, { basePeriod: 14, minPeriod: 6, maxPeriod: 28 });
arsi.forEach(({ value }) => {
  console.log(`RSI: ${value.rsi}, Period: ${value.effectivePeriod}`);
});

// Bollinger Bands with kurtosis-adapted multiplier
const abb = adaptiveBollinger(candles, { period: 20 });

Strategy Robustness Score

import { quickRobustnessScore, runBacktest } from 'trendcraft';

const result = runBacktest(candles, entry, exit, options);
const robustness = quickRobustnessScore(result);

console.log(robustness.grade);          // "A+" to "F"
console.log(robustness.compositeScore); // 0-100
console.log(robustness.recommendations); // ["Strategy passes all robustness checks..."]

Pairs Trading

import { analyzePair } from 'trendcraft';

const result = analyzePair(
  candlesA.map(c => ({ time: c.time, value: c.close })),
  candlesB.map(c => ({ time: c.time, value: c.close })),
);

if (result.cointegration.isCointegrated) {
  console.log(`Hedge ratio: ${result.cointegration.hedgeRatio}`);
  console.log(`Half-life: ${result.meanReversion.halfLife} bars`);
  console.log(`Signals: ${result.signals.length}`);
}

Cross-Asset Correlation

import { analyzeCorrelation } from 'trendcraft';

const analysis = analyzeCorrelation(
  seriesSPY.map(c => ({ time: c.time, value: c.close })),
  seriesQQQ.map(c => ({ time: c.time, value: c.close })),
  { window: 60 },
);

console.log(`Current regime: ${analysis.summary.currentRegime}`);
console.log(`Lead-lag: ${analysis.leadLag.assessment}`);
console.log(`Divergences: ${analysis.divergences.length}`);

CLI Tools

TrendCraft provides two CLI tools for screening and backtesting.

Screening CLI

Screen multiple stocks against entry/exit conditions.

# Basic usage
npx trendcraft-screen ./data --entry "goldenCross"

# With entry and exit conditions
npx trendcraft-screen ./data --entry "goldenCross,volumeAnomaly" --exit "deadCross"

# Filter by ATR% (volatility)
npx trendcraft-screen ./data --entry "perfectOrderBullish" --min-atr 2.3

# JSON output
npx trendcraft-screen ./data --output json > results.json

# Show all stocks (not just signals)
npx trendcraft-screen ./data --all

# List available conditions
npx trendcraft-screen --list

# Specify minimum data points required
npx trendcraft-screen ./data --entry "goldenCross" --min-data 200

Output Example (table):

====================================================================
Stock Screening Results - 2025-12-25
====================================================================

Criteria: goldenCross + volumeAnomaly

Summary:
  Total Files: 19
  Entry Signals: 3
  Exit Signals: 1

--------------------------------------------------------------------
| Ticker       | Signal |     Price |    ATR% |   RSI | Vol Ratio |
--------------------------------------------------------------------
| 6920.T       | ENTRY  |     32500 |   4.52% |    38 |      2.31 |
| 4755.T       | ENTRY  |       850 |   3.21% |    42 |      1.87 |
| 6758.T       | EXIT   |     13200 |   2.34% |    68 |      1.12 |
--------------------------------------------------------------------

Backtest CLI

Run backtests on single files or compare across multiple files.

# Single file (detailed output)
npx trendcraft-backtest ./data/6758.T.csv --entry "goldenCross" --exit "deadCross"

# Multiple files (comparison table)
npx trendcraft-backtest ./data --entry "perfectOrderActiveBullish" --exit "perfectOrderCollapsed"

# With risk management
npx trendcraft-backtest ./data/6758.T.csv \
  --entry "goldenCross,volumeAnomaly" \
  --exit "deadCross" \
  --stop-loss 5 --take-profit 10

# Show individual trades
npx trendcraft-backtest ./data/6758.T.csv --trades

# JSON output
npx trendcraft-backtest ./data --output json > results.json

# List available conditions
npx trendcraft-backtest --list

# Custom capital and commission
npx trendcraft-backtest ./data/6758.T.csv \
  --entry "goldenCross" --exit "deadCross" \
  --capital 500000 --commission 500

# With trailing stop
npx trendcraft-backtest ./data/6758.T.csv \
  --entry "goldenCross" --exit "deadCross" \
  --trailing-stop 3

Single File Output:

======================================================================
Backtest Result: 6758.T
======================================================================

Performance Summary:
  Total Return:     34.48%
  Trade Count:      311
  Win Rate:         34.4%
  Profit Factor:    1.01
  Max Drawdown:     86.88%
  Sharpe Ratio:     1.130

Capital:
  Initial:          1,000,000
  Final:            1,344,761

Multiple Files Output:

Summary:
  Total Stocks:     19
  Profitable:       7 (37%)
  Average Return:   417.90%

--------------------------------------------------------------------------------------------------------------
| Ticker       | Trades |  WinRate |     Return |    MaxDD |     PF |  Sharpe |
--------------------------------------------------------------------------------------------------------------
| 9984.T       |    166 |    41.0% |   7162.68% |   68.78% |   1.42 |   3.090 |
| 6920.T       |     83 |    37.4% |    818.02% |   73.30% |   1.74 |   3.410 |
| 6758.T       |    311 |    34.4% |     34.48% |   86.88% |   1.01 |   1.130 |
--------------------------------------------------------------------------------------------------------------

Available Condition Presets

Both CLI tools support these preset conditions:

| Category | Conditions | |----------|-----------| | Moving Average | goldenCross, deadCross, goldenCross25_75, deadCross25_75 | | Validated Cross | validatedGoldenCross, validatedDeadCross | | RSI | rsiBelow30, rsiBelow40, rsiAbove60, rsiAbove70 | | MACD | macdCrossUp, macdCrossDown | | Perfect Order | perfectOrderBullish, perfectOrderBearish, perfectOrderCollapsed, perfectOrderActiveBullish, perfectOrderActiveBearish, perfectOrderBullishConfirmed, perfectOrderBearishConfirmed, perfectOrderConfirmationFormed, perfectOrderBreakdown, perfectOrderMaCollapsed, perfectOrderPreBullish, perfectOrderPreBearish, perfectOrderPullbackEntry, perfectOrderPullbackSellEntry | | PO Extended | poPlusEntry, pbEntry, poPlusPbEntry | | Price | priceAboveSma25, priceBelowSma25, priceDroppedAtr | | Bollinger | bollingerBreakoutUp, bollingerBreakoutDown | | Stochastics | stochBelow20, stochAbove80, stochCrossUp, stochCrossDown | | DMI/ADX | dmiBullish, dmiBearish, adxStrong | | Volume | volumeAnomaly, volumeAbove1_5x, volumeAbove2x, volumeConfirmsTrend | | Volume Extended | volumeExtreme, volumeDivergence, bullishVolumeDivergence, bearishVolumeDivergence, volumeTrendConfidence, nearPoc, inValueArea, breakoutVah, breakdownVal, priceAbovePoc, priceBelowPoc, cmfAbove, cmfBelow, obvRising, obvFalling, obvCrossUp, obvCrossDown | | Range | rangeBreakout, rangeConfirmed, inRangeBound, tightRange | | Range Extended | rangeForming, breakoutRiskUp, breakoutRiskDown, rangeScoreAbove | | Volatility | atrPercentAbove2_3, atrPercentAbove3 | | Volatility Regime | volatilityExpanding, volatilityContracting, volatilityAbove, volatilityBelow, atrPercentileAbove, atrPercentileBelow, atrPercentBelow | | Patterns | anyBullishPattern, anyBearishPattern, doubleTopDetected, doubleBottomDetected, headShouldersDetected, inverseHeadShouldersDetected, cupHandleDetected | | SMC | priceAtBullishOrderBlock, priceAtBearishOrderBlock, orderBlockCreated, liquiditySweepDetected, liquiditySweepRecovered, hasRecentSweeps |

See API Reference for full condition documentation with parameters and examples.

Cookbook

See Cookbook for practical recipes including:

• Golden Cross + RSI backtest
• Trend following with ADX + volume
• Mean reversion with Bollinger Bands
• Multi-timeframe strategies
• ATR risk management + position sizing
• Parameter optimization with walk-forward validation
• Real-time streaming pipelines
• Scoring-based entry strategies
• Stock screening
• Scaled entry with partial take profit
• Signal explainability for debugging strategies
• Composable indicator pipelines
• Alpha decay monitoring
• Pairs trading with cointegration
• Cross-asset correlation analysis

API Reference

See API Documentation for detailed API reference.

Data Format

Input: Candle Data

interface Candle {
  time: number | string | Date;  // Timestamp, date string, or Date object
  open: number;
  high: number;
  low: number;
  close: number;
  volume: number;
}

Output: Indicator Values

interface IndicatorValue<T> {
  time: number;   // Unix timestamp (milliseconds)
  value: T;       // Indicator value (type varies by indicator)
}

type Series<T> = IndicatorValue<T>[];

License

MIT