This presentation is a presentation for the Army Conference on Applied Statistics. The presentation covers a brief introduction to land warfare problems, and devises a methodology using Bayes Theorem to estimate parameters of interest. Two examples are given, a simple one using independent Bernoulli Trials, and a more complex one using correlated Red and Blue casualty data in a Loss Exchange Ratio and a hierarchical model. The presentation demonstrates that the Bayesian approach is successful in both examples at reducing the variance of the estimated parameters, potentially reducing the cost of devising a complex test program. The presentation concludes with suggested next steps applicable to the Army Ground Combat Vehicle program.
Suggested Citation
Wilson, Alyson, Lee Dewald, Robert Holcomb, and Samuel Parry. A Bayesian Approach to Evaluation of Land Warfare Systems. IDA Document NS D-4711. Alexandria, VA: Institute for Defense Analyses, 2012.