@umn-latis/simple-smoothing-spline

Simple Smoothing Spline

Fit a smoothing spline to collection of data points

Usage

const data = [
  { x: 1, y: 0.5 },
  { x: 2, y: 3 },
  { x: 3, y: 8.5 },
  { x: 4, y: 20 },
  { x: 1, y: 1 },
  { x: 2, y: 5 },
  { x: 3, y: 11 },
  { x: 4, y: 15 },
];

const spline = await simpleSmoothingSpline(data, { lambda: 1000 });

// spline.points is a collection of {x, y} values
// between the min and max the x values in the data set
graphItWithYourOwnFunction(spline.points);

// spline.fn is a function that can return a y for a given x
const myY = spline.fn(2.5);
// 6.25

Installation

Using NPM or Yarn

npm install @umn-latis/simple-smoothing-spline

yarn add @umn-latis/simple-smoothing-spline

With a <script> tag

Added this before your main script:

<!-- index.html -->
<script
  src="https://unpkg.com/@umn-latis/simple-smoothing-spline/dist/index.umd.min.js"
  async
></script>

Then, in your scripts, you can call a global simpleSmoothingSpline() function.

API

async simpleSmoothingSpline(data, opts)

Parameters:

• data - an array of data points in the form of {x: 1, y: 2}.

• opts.lambda = 1000 - lambda parameter for Ridge regression. This is the tuning parameter for the regression function. The higher the lambda, the smoother the spline will be. By default, this is 1000.

• opts.type = smooth (default) | cubic - type of spline to use for regression.

  Example:

  const cubicSpline = await simpleSmoothingSpline(data, { type: "cubic" });

Returns a Promise for::

• spline.points - An array of {x, y} points on the smoothing spline in the range of the data (between min(x) and max(x)).

• spline.fn - A function to get an arbitrary f(x) for a given x.

  Example:

  const spline = await simpleSmoothingSpline(data, { lambda: 2 ** 8 });
  const y = spline.fn(3);
  // y is value on the spline when x = 3

Example

An example using Plotly JS can be found at https://umn-latis.github.io/simple-smoothing-spline

About

• Background notes and references
• Maintainers
• Contributors

License

MIT