Confidence Intervals for the Product of Two Binomial Parameters

Abstract

Many more-or-lese reasonable solutions to the problem of statistical estimation of the product P1P. of two binomial parameters by confidence intervals can be given. After specializing to the case where P 1 and P. are much less than 1 and to estimation by "one-sided" intervals it is shown that a unique solution is obtained when one assumes a certain set of inequalities and then requires the intervals to be as short as possible. Tables of intervals for 90 and 95 per cent confidence levels are presented based on reasonable sets of inequalities and on a Poisson approximation to the binomial.

Approximate Confidence Limits for Complex Systems With Exponential Time Until Failure of the Components

“…), then the probability of the event specified in (1.4) is Sampford [11]. Also in this connection see the discussion by Cohen in [2].…”

Maximum Likelihood Estimation for a Class of Multinomial Distributions Arising in Reliability.

“…Of particular importance are 0 problems in which several p r 's are very large and/or several P t 's are very small. As Buehler (1957) has noted , such problems occur with considerable frequency in reliability experiments . It is interesting k to note that the method proposed here for the case 11 n~ > 0 tends to 0 work nicely for problems in which the integers with nonzero observed frequencies are consecutive --precisely the expected data configuration for the problem of interest.…”

“…The problem does not seem to benefit , however, from either of the approach es used in these studies . An estimation pr oblem f or th is model with k a 2 was considered by Buehler (1957).…”

Bounds for Optimal Confidence Limits for Series Systems

“…Schafer, and Singpurvslla (1974). Historically, the first significant work on this problem was produced by Buehler (1957). However.…”