node-statistics v1.0.4

Node-statistics

A JavaScript implementation of mean/variance, and a collection of regression methods for finding the trend of data using R language.

Install

npm install node-statistics

Requirement

Must able to run R in shell, visit CRAM to have R installed.

Usage

nodeStat.mean(array, cb)

The first argument is an array of numeric value, and cb is a callback funciton of err and result

return:

	{ mean: ...,
	  median: ...,
	  min: ...,
	  max: ...,
	  sum: ...,
	  variance: ...,
	  stdDev: ...,
	  quantile: { '0':..., '25':..., '50':..., '75': ..., '100':... } 
	 }

nodeStat.regression(method, array, array, cb)

The first argument is an string represents the method of regression, supported reggresion are linear, quadratic, cubic, log. The second and third argument are two array of numeric value, it is x and y value in regression respectfully. The last argument is a callbcak function.

return:

	{
    residuals: [...],
    p-value: ...,
    sigma: ...,
    coefficients: {
    variable : { Estimate: ...,
                 StdErr  : ...,
                 t-value : ...,
                 PrGtT   : ... }
               }
    }