@turf/moran-index

moranIndex

Moran's I measures patterns of attribute values associated with features. The method reveal whether similar values tend to occur near each other, or whether high or low values are interspersed.

Moran's I > 0 means a clusterd pattern. Moran's I < 0 means a dispersed pattern. Moran's I = 0 means a random pattern.

In order to test the significance of the result. The z score is calculated. A positive enough z-score (ex. >1.96) indicates clustering, while a negative enough z-score (ex. <-1.96) indicates a dispersed pattern.

the z-score can be calculated based on a normal or random assumption.

Bibliography*

1. Moran's I

2. pysal

3. Andy Mitchell, The ESRI Guide to GIS Analysis Volume 2: Spatial Measurements & Statistics.

Parameters

• fc FeatureCollection<any>
• options Object
  • options.inputField string the property name, must contain numeric values
  • options.threshold number the distance threshold (optional, default 100000)
  • options.p number the Minkowski p-norm distance parameter (optional, default 2)
  • options.binary boolean whether transfrom the distance to binary (optional, default false)
  • options.alpha number the distance decay parameter (optional, default -1)
  • options.standardization boolean wheter row standardization the distance (optional, default true)

Examples

const bbox = [-65, 40, -63, 42];
const dataset = turf.randomPoint(100, { bbox: bbox });

const result = moranIndex(pts, {
  inputField: 'CRIME',
});

Returns MoranIndex

mean

get mean of a list

Parameters

• y Array<number>

Returns number

variance

get variance of a list

Parameters

• y Array<number>

Returns number

MoranIndex

Type: Object

Properties

• moranIndex number the moran's Index of the observed feature set
• expectedMoranIndex number the moran's Index of the random distribution
• stdNorm number the standard devitaion of the random distribution
• zNorm number the z-score of the observe samples with regard to the random distribution

---

This module is part of the Turfjs project, an open source module collection dedicated to geographic algorithms. It is maintained in the Turfjs/turf repository, where you can create PRs and issues.

Installation

Install this module individually:

$ npm install @turf/moran-index

Or install the Turf module that includes it as a function:

$ npm install @turf/turf