Topic
Sequential probability ratio test
About: Sequential probability ratio test is a research topic. Over the lifetime, 1248 publications have been published within this topic receiving 22355 citations.
Papers published on a yearly basis
Papers
More filters
••
01 Jan 2003TL;DR: It is demonstrated that, for these simulated long-term monitoring scenarios, decisions to issue an alarm when the measured count rate equals the threshold count rate are made 3-5 times faster using the SPRT than with the SIT.
Abstract: Among the possible decision-making algorithms for sequentially-acquired radiation sensor data is the Sequential Probability Ratio Test (SPRT). The suitability of the SPRT for long-term monitoring applications is discussed, and the decision-making performance of the SPRT is compared to that of the commonly used single-interval test (SIT). The analysis spans a wide range of signal and background count rates so that results are applicable to sensors of all sizes operating in different ambient conditions, with a spectrum of alarm thresholds. It is demonstrated that, for these simulated long-term monitoring scenarios, decisions to issue an alarm when the measured count rate equals the threshold count rate are made 3-5 times faster using the SPRT than with the SIT. The ability of the SPRT to provide an "all-clear" indication and the need for SPRT truncation strategies to limit decision times when the measured count rate falls between background and the specified threshold are also discussed. Under an early termination scenario, it is shown that a truncated SPRT retains a higher probability of detection.
52 citations
••
TL;DR: In this paper, the authors consider hypothesis testing problems in which a nuisance parameter is present only under the alternative hypothesis, and they reparameterize the testing problem to one for which an exact small sample test can be construct ed using existing hypothesis testing procedures.
Abstract: The authors consider hypothesis testing problems in which a nuisance parameter is present only under the alternative hypothesis. Standard asymptotic tests, such as likelihood ratio, Lagrange multiplier and Wald tests, are difficult to apply because o f problems incurred in obtaining their asymptotic distributions. To overcome this difficulty, the authors reparameterize the testing problem to one for which an exact small sample test can be construct ed using existing hypothesis testing procedures. The reparameterization technique is applied to two examples from the econometrics literatur e, and an empirical power comparison shows that their test has better power properties than tests previously proposed in the literature. Further, p-values for their test can be computed. in O(n) operations so the test can be implemented efficiently. Copyright 1993 by MIT Press.
51 citations
••
08 Jan 2006TL;DR: A new statistical solution method is presented that can bound the probability of error under any circumstances by sometimes reporting undecided results, and is presented as a framework for expressing correctness guarantees of model-checking algorithms.
Abstract: We introduce a framework for expressing correctness guarantees of model-checking algorithms. The framework allows us to qualitatively compare different solution techniques for probabilistic model checking, both techniques based on statistical sampling and numerical computation of probability estimates. We provide several new insights into the relative merits of the different approaches. In addition, we present a new statistical solution method that can bound the probability of error under any circumstances by sometimes reporting undecided results. Previous statistical solution methods could only bound the probability of error outside of an “indifference region.”
50 citations
••
TL;DR: The efficiency of the Wald sequential probability ratio test relative to the best competing fixed sample procedure for testing H 0:θ = θ 0 versus H 1: θ = ǫ 1 (θ 0 0) was shown in this paper.
Abstract: The efficiency (measured in terms of ratio of average sample size to fixed sample size) of the Wald sequential probability ratio test relative to the best competing fixed sample procedure for testing H 0:θ = θ0 versus H 1:θ = θ1 (θ0 0) this limiting relative efficiency is equal to (θ1 – θ0)/(4|θ1+θ0 − 2θ|). The practical implications of this result are discussed.
49 citations