Advanced Scientific Analysis

SciChart Engine includes a comprehensive set of mathematical functions for professional scientific analysis. All functions are optimized for large datasets and are available directly from the chart instance.

Interactive Demo

Experiment with real-time analysis functions:

📊 0 pts🚀 0 FPS

Transforms

Analysis

Las transformaciones son acumulativas: activa varias para combinarlas.

---

Available Functions

📈 Savitzky-Golay (Smart Smoothing)

The Savitzky-Golay filter fits a local polynomial to a window of points, removing noise while preserving the shape, width, and height of peaks.

import { PluginAnalysis } from 'scichart-engine';

// Using the analysis plugin
const analysis = chart.getPluginAPI('scichart-analysis');
const smoothed = analysis.savitzkyGolay(rawSignal, 11, 3);

Parameters:

• data: Data array to smooth.
• windowSize: Window size (must be odd).
• polynomialOrder: Polynomial order (typically 2-4).

---

📉 Baseline Correction

Eliminates linear drift or signal shifts to isolate true peaks.

import { subtractBaseline } from 'scichart-engine';

// Subtract linear baseline between x=0 and x=10
const corrected = subtractBaseline(x, y, 0, 10);

Typical Use:

• Electrochemical signals with capacitive drift.
• Spectra with background tilt.
• Chromatograms with varying baseline.

---

🎯 Peak Detection

Automatically detects local peaks with prominence filtering.

const peaks = chart.analysis.detectPeaks(x, y, {
  minProminence: 0.5,  // Minimum prominence
  type: 'max'          // 'max', 'min', or 'both'
});

// Result
peaks.forEach(peak => {
  console.log(`Peak at x=${peak.x}, y=${peak.y}, prominence=${peak.prominence}`);
});

Options:

• minProminence: Minimum height relative to neighboring valleys.
• type: Type of extrema to detect.

---

∫ Numerical Integration

Calculates the area under the curve using the trapezoidal rule.

// Total Area
const totalArea = chart.analysis.integrate(x, y);

// Area in specific range
const partialArea = chart.analysis.integrate(x, y, { xMin: 2, xMax: 5 });

Applications:

• Total charge (Q = ∫I dt) in electrochemistry.
• Concentration in chromatography.
• Energy in power signals.

---

📐 Numerical Derivatives

Calculates the first or second derivative of the data.

// First derivative
const dy = chart.analysis.derivative(x, y, 1);

// Second derivative with prior smoothing
const d2y = chart.analysis.derivative(x, y, 2, 5);

Uses:

• Finding inflection points.
• Detecting slope changes.
• Titration analysis.

---

🔍 LTTB Downsampling

Reduces the number of points while preserving visual shape (Largest-Triangle-Three-Buckets).

// Reduce 100,000 points to 1,000
const downsampled = chart.analysis.downsampleLTTB(x, y, 1000);

chart.addSeries({
  id: 'optimized',
  type: 'line',
  data: downsampled
});

Benefits:

• Extreme performance with large datasets.
• Preserves peaks and valleys (unlike averaging).
• Identical visualization with 100x fewer points.

---

📏 Delta Tool (Measurement Tool)

Measures distances, deltas, and slopes between two points on the chart.

// Enable Delta Tool (auto-loaded)
chart.setMode('delta');

// Listen for measurements
chart.on('measure', (measurement) => {
  console.log(`ΔX: ${measurement.deltaX}`);
  console.log(`ΔY: ${measurement.deltaY}`);
  console.log(`Slope: ${measurement.slope}`);
});

---

FFT and Digital Filters

Fourier Transform

// Calculate frequency spectrum
const spectrum = chart.analysis.fft(timeDomainData);
console.log(spectrum.magnitude); // Amplitude
console.log(spectrum.phase);     // Phase

// Inverse transform
const reconstructed = chart.analysis.ifft(spectrum);

FIR Filters

// Low-pass filter at 100 Hz with sample rate 1000 Hz
const filtered = chart.analysis.lowPassFilter(data, 100, 1000);

// High-pass filter
const highPassed = chart.analysis.highPassFilter(data, 50, 1000);

// Band-pass filter
const bandPassed = chart.analysis.bandPassFilter(data, 50, 200, 1000);

Butterworth Filter

// Order 4 Butterworth with cutoff frequency 0.1 (normalized)
const filtered = chart.analysis.butterworth(data, 4, 0.1);

---

Financial Indicators

// Simple Moving Average (20 periods)
const ma20 = chart.analysis.sma(closeData, 20);

// Exponential Moving Average
const ema12 = chart.analysis.ema(closeData, 12);

// MACD
const macdResult = chart.analysis.macd(closeData, 12, 26, 9);

// RSI
const rsiValues = chart.analysis.rsi(closeData, 14);

// Bollinger Bands
const bands = chart.analysis.bollingerBands(closeData, 20, 2);

---

Statistics and Anomaly Detection

// Cross-correlation between two signals
const correlation = chart.analysis.crossCorrelation(signal1, signal2);

// Anomaly detection with Z-score
const anomalies = chart.analysis.detectAnomalies(data, {
  method: 'zscore',
  threshold: 3
});

// Two-sample T-test
const result = chart.analysis.tTest(sample1, sample2);
console.log(`p-value: ${result.pValue}, significant: ${result.significant}`);

---

Accessing from the Chart

Most analysis functions are available directly via the PluginAnalysis (auto-loaded in ChartCore):

const chart = createChart({ container });

// Access analysis via plugin API
const analysis = chart.getPluginAPI('scichart-analysis');

analysis.savitzkyGolay(data, 11, 3);
analysis.fft(signal);
analysis.sma(prices, 20);

This ensures you can perform complex analysis without additional imports, with direct access to chart features.