Curve Fitting Demo

SciChart Engine provides a powerful and flexible curve fitting system. You can automatically calculate and render trend lines for any existing series using various mathematical models.

Interactive Example

The following demo shows two datasets:

1. Blue dots: A linear calibration plot ($y = ax + b$) with random noise, fitted with a linear regression.
2. Red squares: A 3rd-degree polynomial dataset with noise, fitted with a cubic polynomial regression.

📊 0 points 🚀 0 FPS

Scroll to zoom • Drag to pan • Linear and Polynomial Regression

How it Works

The addFitLine method performs the mathematical regression on the GPU-ready data of a series and automatically:

• Generates a smooth trend line series.
• Styles it as a dashed line with matched colors.
• Adds an interactive text annotation with the equation and the $R^2$ value.

Configuration

// Add a linear fit
chart.addFitLine('my-series', 'linear', { 
  precision: 3 
});

// Add a high-order polynomial fit
chart.addFitLine('data-points', 'polynomial', { 
  degree: 3, 
  precision: 2 
});

// Other supported types: 'exponential', 'logarithmic', 'power'
chart.addFitLine('growth-data', 'exponential');

Features

• Linear Regression: Best fit line using least squares.
• Polynomial Fit: Supports any degree (typically 2-5).
• Non-linear Models: Logarithmic, Exponential, and Power fits.
• Automatic Labels: Equation and $R^2$ are automatically updated and placed.
• Dashed Lines: Uses the new lineDash styling property for better visualization.

Use Cases

• Electrochemistry: Calibration curves, Tafel plots, and kinetic analysis.
• Physics: Data modeling and trend identification.
• Economics: Forecasts and growth modeling.
• Quality Control: Monitoring drift and stability.