qsd: A normal model

In qle: Simulation-Based Quasi-Likelihood Estimation

Description

A statistical model of random numbers

Usage

data(normal)

Format

A list object named 'qsd' of class QLmodel with additional elements

• simfn simulation function

• sim simulation results at design points, class 'simQL'

• OPT result from call to estimation function qle

• QS quasi-scoring iteration results after initial approximation

Details

This is a pedagogic example of a simulated data set for quasi-likelihood estimation using normally distributed random numbers. The model outcome is a vector of summary statistics, that is, simply the median and mean average deviation of n=10 random numbers, which is evaluated at the model parameter θ=(μ,σ) with mean μ and standard deviation σ as the parameters of the normal distribution. We estimate the model parameter given a specific "observation" of those summary statistics. Clearly, maximum likelihood estimation would be the method of first choice if we had a real sample of observations. However, this example is used to demonstrate the basic workflow of estimating the model parameter. We use this model as a standard example in the package documentation.

Author(s)

M. Baaske

qle documentation built on May 2, 2019, 5:26 p.m.