Weibull Parameter estimation using Apache Commons Math

Question

I need to estimate the shape and scale parameters of a Weibull distribution from a sample data using MLE. I checked the WeibullDistribution class of Apache commons Math but it doesn't have such functionality. Any suggestion? I need to use the class in my java application.

Statfarad

Did you check this project: github.com/SquareBracketAssociates/ArchiveOONumericalMethods/…. It's possible to create a WeibullDistribution from a Histogram (you have to provide data here). And following this: stats.stackexchange.com/questions/8960/… you could refactor the method to do exactly what you need (using MLE). – Despiteful

Answer 1

There are no methods that calculate directly MLE estimations for Weibull distribution in Apache Commons Math, but since you know the estimating equations for scale and shape, you can use the method solve() of the class NewtonRaphsonSolver of Apache Commons Math to compute the estimation of the shape parameter, which use Newton-Rhapson approach, whereas for the scale I think you can just use the evaluate() method of the Sum class after you have the result for shape.

Bethune