Enter your mobile number or email address below and we'll send you a link to download the free Kindle App. Then you can start reading Kindle books on your smartphone, tablet, or computer - no Kindle device required.
To get the free app, enter your mobile phone number.
Statistical measures of accuracy for riflemen and missile engineers Unknown Binding – 1991
|New from||Used from|
Author interviews, book reviews, editors picks, and more. Read it now
Top customer reviews
There was a problem filtering reviews right now. Please try again later.
Reading any second year statistics textbook will reveal that arrival times and locations (e.g. bullet impacts) are realizations of a discretely deterministic process. Modern statistics dictates Poisson process is to be preferred. My experience tells me that the best distributional assumption is going to be a log-Gaussian Cox process of some sort. Not only do these start in the right place (i.e. stochastic events), but they can be tweaked to describe events seen in the data: +Defects in rifling producing patterned errors can be addressed with a stationary cluster process. +Drift in bullet mean location due to change in barrel temperature can be addressed by adding some attractive Markov-type process to the generating function. +Changes in accuracy due to barrel temperature can be addressed by adding a time-varying parameter. +The interaction of drift in mean point of impact and increased accuracy from a warm barrel can be handled by adding a Hawkes process.
This was an important booklet. And it is still useful for hobbyists with some background in statistics. But if you're actually good at statistics, it's just ... annoying. If you want to perform serious statistical analysis on bullet arrival patterns, you'd do far better to start with any of the standard textbooks on spatial statistics (esp. Moller & Waagepetersen "Statistical Inference and Simulation for Spatial Point Processes").