Correlate Method

The correlate() method calculates the Pearson correlation coefficient between two numeric fields in the collection. This method helps identify linear relationships between variables, returning a value between -1 and 1.

Basic Syntax

correlate<K extends keyof T>(key1: K, key2: K): number

Examples

Basic Usage

import { collect } from 'ts-collect'

const data = collect([
  { x: 1, y: 2 },
  { x: 2, y: 4 },
  { x: 3, y: 6 },
  { x: 4, y: 8 }
])

const correlation = data.correlate('x', 'y')
console.log(correlation) // 1.0 (perfect positive correlation)

Working with Real Data

interface MarketData {
  date: string
  price: number
  volume: number
  volatility: number
}

const marketData = collect<MarketData>([
  { date: '2024-01-01', price: 100, volume: 1000, volatility: 0.15 },
  { date: '2024-01-02', price: 102, volume: 1200, volatility: 0.18 },
  { date: '2024-01-03', price: 98, volume: 1500, volatility: 0.22 }
])

// Check correlation between volume and volatility
const volCorrelation = marketData.correlate('volume', 'volatility')

Real-world Example: E-commerce Analytics

interface SalesMetrics {
  date: string
  advertising: number
  visitors: number
  sales: number
  averageOrderValue: number
  conversionRate: number
}

class SalesAnalyzer {
  private metrics: Collection<SalesMetrics>

  constructor(metrics: SalesMetrics[]) {
    this.metrics = collect(metrics)
  }

  analyzeMetricCorrelations() {
    return {
      advertisingToSales: this.metrics.correlate('advertising', 'sales'),
      visitorsToSales: this.metrics.correlate('visitors', 'sales'),
      visitorToConversion: this.metrics.correlate('visitors', 'conversionRate'),
      salesToOrderValue: this.metrics.correlate('sales', 'averageOrderValue')
    }
  }

  findStrongestSalesDriver(): string {
    const correlations = [
      { metric: 'Advertising', value: this.metrics.correlate('advertising', 'sales') },
      { metric: 'Visitors', value: this.metrics.correlate('visitors', 'sales') },
      { metric: 'Conversion Rate', value: this.metrics.correlate('conversionRate', 'sales') }
    ]

    return collect(correlations)
      .sortBy('value', 'desc')
      .first()
      ?.metric || 'No significant driver found'
  }

  getCorrelationStrength(key1: keyof SalesMetrics, key2: keyof SalesMetrics): string {
    const correlation = Math.abs(this.metrics.correlate(key1, key2))

    if (correlation >= 0.7) return 'Strong'
    if (correlation >= 0.3) return 'Moderate'
    return 'Weak'
  }
}

// Usage
const analyzer = new SalesAnalyzer([
  {
    date: '2024-01-01',
    advertising: 1000,
    visitors: 5000,
    sales: 15000,
    averageOrderValue: 75,
    conversionRate: 0.02
  },
  // ... more data
])

const analysis = analyzer.analyzeMetricCorrelations()

Type Safety

interface DataPoint {
  id: number
  x: number
  y: number
  label: string
}

const points = collect<DataPoint>([
  { id: 1, x: 10, y: 20, label: 'A' },
  { id: 2, x: 15, y: 25, label: 'B' }
])

// Type-safe correlation
const correlation: number = points.correlate('x', 'y') // ✓ Valid

// TypeScript enforces numeric fields
// points.correlate('x', 'label') // ✗ TypeScript error
// points.correlate('label', 'y') // ✗ TypeScript error

Return Value

• Returns a number between -1 and 1 where:
  • 1: Perfect positive correlation
  • -1: Perfect negative correlation
  • 0: No correlation
• NaN if data is insufficient or invalid
• Undefined values are skipped
• Non-numeric values cause TypeError
• Requires at least 2 points
• Same field correlation returns 1

Common Use Cases

1. Sales Analysis

• Marketing impact analysis
• Conversion optimization
• Price sensitivity studies
• Performance metrics
• Customer behavior analysis

2. Performance Metrics

• Resource utilization
• System metrics correlation
• Performance indicators
• Load analysis
• Capacity planning

3. Marketing Analysis

• Campaign effectiveness
• Channel performance
• Budget optimization
• ROI analysis
• Attribution modeling

4. Customer Insights

• Behavior patterns
• Purchase correlations
• Engagement metrics
• Retention analysis
• Value drivers

5. Risk Analysis

• Market indicators
• Risk factors
• Pattern detection
• Trend analysis
• Predictive metrics