Book reviews

Third World Military Expenditure: Determinants and Implications. By Robert McKinlay. Frances Pinter, London (1989)

The UK Defence Industrial Base: Development and Future Policy Options. By Trevor Taylor and Keith Hayward. Brassey's, London, for Royal United Services Institute (1989), ISBN 0-08-036713-5, £22.50

Mutiny. By Lawrence James. Buchan & Enright, London (1987), ISBN 0-907675-70-0, £12.95; Scapegoat! Famous Courts Martial. By John Harris. Severn House, London (1988), ISBN 0-7278-2103-2, £12.95; In Glass Houses. By Robert Boyes. Military Provost Staff Corps Association, Colchester (1986), ISBN 0-9513467-0-9, £6.50 (paperback)

The Nuclear Weapons World: Who, How and Where. Edited by Patrick Burke. Frances Pinter, London (1988), ISBN 086187-705-5, £50.00

Merchants of Treason—America's Secrets for Sale. By Thomas B. Allen and Norman Polmar. Robert Hale, London (1988), ISBN 0-7090-3543-8, £14.95 ($21.95); Intelligence and Intelligence Policy in a Democratic Society. Edited by Stephen J. Cimbala. Transnational Publishers, Dobbs Ferry, NY (1987), ISBN 0-941320-44-8, $37.50; Catching Spies—Principles and Practices of Counterespionage. By H. H. A. Cooper and Lawrence J. Redlinger. Paladin Press, Boulder, CO (1988), ISBN 0-87364-466-2, $24.95

The BattleshipRoyal Sovereignand Her Sister Ships. By Peter C. Smith. William Kimber, Wellingborough (1988), ISBN 0-7183-0704-6, £12.95; Air Power at Sea, 1945 to Today. By John Winton. Sidgwick & Jackson, London (1987), £9.95     

Book reviews

British Military History: a Supplement to Robin Higham's Guide to the Sources.. Edited by Gerald Jordan. Garland, New York (1988), xii + 586 pp., $75.00

Red Duster at War. By John Slader. William Kimber, London (1988), 352 pp., £15.50

Armed Services and Society. By Martin Edmonds. Leicester University Press, Leicester (1988), 226 pp., £25.00     

A sequential bayes procedure for reliability demonstration

A common problem in life testing is to demonstrate that the mean time to failure, θ, exceeds some minimum acceptable value, say θ₁, with a given confidence coefficient γ. When this is true, it is said that “θ₁ has been demonstrated with a confidence γ”. In this paper a Sequential Bayes Procedure (SBP) for demonstrating (by means of. a probability statement) that θ exceeds θ₁ is presented. The SBP differs from the classical procedure in the sense that a prior distribution is assumed on the parameter θ, calling for a Bayesian approach. The procedure is based on the sequence of statistics.     

Interception in a network

This paper presents an algorithm for determining where to place intercepting units in order to maximize the probability of preventing an opposing force from proceeding from one particular node in an undirected network to another. The usual gaming assumptions are invoked; namely, the strategy for placing the units is known to the opponent and he will choose a path through the network which, based on this knowledge, maximizes his probability of successful traverse. As given quantities, the model requires a list of the arcs and nodes of the network, the number of intercepting units available to stop the opposing force, and the probabilities for stopping the opposition at the arcs and nodes as functions of the number of intercepting units placed there. From these quantities, the algorithm calculates the probabilities for placing the unit at the arcs and nodes when one intercepting unit is available, and the expected numbers of units to place at the arcs and nodes when multiple intercepting units are available.     

Estimating value in a uniform auction

Consider an auction in which increasing bids are made in sequence on an object whose value θ is known to each bidder. Suppose n bids are received, and the distribution of each bid is conditionally uniform. More specifically, suppose the first bid X₁ is uniformly distributed on [0, θ], and the i^th bid is uniformly distributed on [X_i?1, θ] for i = 2, …?, n. A scenario in which this auction model is appropriate is described. We assume that the value θ is un known to the statistician and must be esimated from the sample X₁, X₂, …?, X_n. The best linear unbiased estimate of θ is derived. The invariance of the estimation problem under scale transformations in noted, and the best invariant estimation problem under scale transformations is noted, and the best invariant estimate of θ under loss L(θ, a) = [(a/θ) ? 1]² is derived. It is shown that this best invariant estimate has uniformly smaller mean-squared error than the best linear unbiased estimate, and the ratio of the mean-squared errors is estimated from simulation experiments. A Bayesian formulation of the estimation problem is also considered, and a class of Bayes estimates is explicitly derived.     

Refined prediction for linear regression models

Adequate prediction of a response variable using a multiple linear regression model is shown in this article to be related to the presence of multicollinearities among the predictor variables. If strong multicollinearities are present in the data, this information can be used to determine when prediction is likely to be accurate. A region of prediction, R, is proposed as a guide for prediction purposes. This region is related to a prediction interval when the matrix of predictor variables is of full column rank, but it can also be used when the sample is undersized. The Gorman-Toman ten-variable data is used to illustrate the effectiveness of the region R.