Analysis of Gunnery Data
Abstract
Data from the gunnery Table VIII exercises in the M1A2 test at Fort Hunter Liggett were used to obtain maximum likelihood estimates of the parameters in a bivariate normal distribution of
impacts of tank main gun rounds. The data are censored; impact
position information is known only for rounds that hit the target
panels. It is argued that much useful information was gained by
observing impact position data, rather than recording only hit-miss
data.