Fitting Expectation to Identify Plausible Explanations
Abstract
The investigation of a model of combustion led to a novel estimation procedure. The a priori knowledge was the equation: impulse is equal to average force multiplied by time duration, or I ft = . The data provided was the mean and standard deviation of the impulse and time. The information requested was the mean and standard deviation of a factor representing average force. The procedure is based on the formula for the variance of a product of random variables. Starting with a simplified model, terms are added until a feasible solution exists. Some of the terms required by the model are not available, and a simulation-based expectation fitting algorithm is used to estimate these unknown values. Through simulation and an assumption about the type of distribution, it is possible to investigate questions that do not have a closed form solution.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jul 01, 2008
- Accession Number
- ADA485308
Entities
People
- Andrew A. Thompson
Organizations
- United States Army Research Laboratory