Showing papers on "Sequential probability ratio test published in 1989"
Book•
01 Sep 1989
TL;DR: In this article, auxiliary signals are used for improving fault detection in the chemical process, and a sequential probability ratio test is used to detect faults in a chemical process with auxiliary signals.
Abstract: Preliminaries.- Sequential probability ratio test.- Auxiliary signals for improving fault detection.- Extension to multiple hypothesis testing.- Modelling and identification of the chemical process.- Fault detection and diagnosis in the chemical process.- Conclusions and further research.
132 citations
TL;DR: In this article, a sequential probability ratio test based on the number of arrivals Xn during the nth service period for M/Ek/l queue is proposed to regulate the traffic intensity.
Abstract: A sequential probability ratio test based on the number of arrivals Xn during the nth service period for M/Ek/l queue is proposed to regulate the traffic intensity. Based on observations Xn, the formulas for operating characteristic and average sample number functions are derived. A numerical example of the implementation of the method is given, and its advantages relative to alternative approaches are discussed.
12 citations
21 Jun 1989
TL;DR: In this article, the authors generalize the sequential probability ratio test (SPRT) of Wald to a distributed system, where each local detector performs an SPRT based on its own observations and communicates its local decision to the global decision maker.
Abstract: In this paper, we generalize the sequential probability ratio test (SPRT) of Wald to a distributed system. In particular, we consider a network of two detectors which are connected in parallel. Each local detector performs an SPRT based on its own observations and communicates its local decision to the global decision maker. The global decision maker combines the local decisions according to a predetermined fusion rule, and decides either to terminate the overall test and accept one of the two hypotheses, or to continue. The global error probabilities are shown to be functions of the local error probabilities and the fusion rule. The global test length is formulated in terms of the local test lengths, and the average global test length is derived. An example is presented for the case of two identical local detectors. The results obtained show clearly that for the same level of performance, the proposed scheme has a shorter average decision time than the single detector case.
11 citations
TL;DR: It is demonstrated how previous work dealing with robust error probability evaluation in the presence of uncertain serial dependency can be applied to sequential tests, and the utility of this extension to Markov-additive process is investigated.
Abstract: The development of an exponential transformation of measure not restricted to independent identically distributed data is presented for a class of semi-Markov processes known as Markov-additive (MA) process. The MA measure transformation is applied to obtain a new form of the Wald identity. To demonstrate the utility of this extension to MA processes, the application of sequential tests with Markov data to error probability performance analysis is investigated. In particular, it is demonstrated how previous work dealing with robust error probability evaluation in the presence of uncertain serial dependency can be applied to sequential tests. >
10 citations
TL;DR: In this paper, the Neyman-Pearson test with fixed level and power is viewed as a Wald test subject to restrictions on the payoff vector, cost function, and prior distribution.
Abstract: The Neyman-Pearson Lemma describes a test for two simple hypotheses that, for a given sample size, is most powerful for its level. It is usually implemented by choosing the smallest sample size that achieves a prespecified power for a fixed level. The Lemma does not describe how to select either the level or the power of the test. In the usual Wald decision-theoretic structure there exists a sampling cost function, an initial prior over the hypothesis space and various payoffs to right/wrong hypothesis selections. The optimal Wald test is a Bayes decision rule that maximizes the expected payoff net of sampling costs. This paper shows that the Wald-optimal test and the Neyman-Pearson test can be the same and how the Neyman-Pearson test, with fixed level and power, can be viewed as a Wald test subject to restrictions on the payoff vector, cost function and prior distribution.
10 citations
TL;DR: In this article, the Beatties procedure for continuous process-monitoring is extended to monitor a process in which the observed variate is normally distributed, which can be used to monitor two-sided variations of the parameter under observation.
Abstract: Beatties procedure for continuous process-monitoring is extended to monitor a process in which the observed variate is normally distributed. This procedure can be used to monitor two-sicled variations of the parameter under observation. Wald's approximating expressions for the sequential probability ratio test are used to obtain an approximate operating characteristic curve for the procedure
3 citations
2 citations
TL;DR: In this paper, an upper bound for the differences of the error probabilities and their Wiener process approximations is derived for the first three moments of the log-likelihood ratio and is seen to be especially useful in the case of contiguous hypotheses.
1 citations
01 Jan 1989
TL;DR: A major feature of the proposed model is that the information accessible to the controller includes proprioceptive information associated with control manipulation in addition to visual information.
Abstract: This paper proposes a model of human operators' failure detection behavior in actively controlling a linear system and passively monitoring signals on the display screen. Estimation and signal detection theories were appliied to the development of this model. Its mechanism of failure detection is expressed by Wald's Sequential Probability Ratio Test, and two autoregressive models are used to predict the values at one step ahead of signals - signals on the display screen and a displacement of the manipulator - from their current values as perceived by the human operator. One of these autoregressive models is for hypothesis H0 where the system is in normal working order, and the other is for hypothesis H1 where the system is out of order. A major feature of the proposed model is that the information accessible to the controller includes proprioceptive information associated with control manipulation in addition to visual information. The detection time and accuracy predicted by this model corresponded well with the data obtained from experiments over a wide range for the controller and monitor.
TL;DR: In this paper, a continuous process monitoring procedure that does not require 100% inspection is proposed. But the procedure is a composite of two cumulative sum procedures and is not suitable for the case of discrete variates under observation.
Abstract: Beattie (1962) devised a continuous process-monitoring procedure that does not require 100% inspection. The procedure is a composite of two cumulative sum procedures. He considered the case in which the variate under observation is discrete. Here we extend Beattie's procedure to monitor a process when the observed variate is normally distributed. Additionally, we show how this procedure can be used to monitor two-sided variations of the parameter under observation. We use Wald's approximating expressions for the sequential probability ratio test to obtain an approximate operating characteristic (OC) curve for the procedure. We also evaluate the accuracy of the OC-curve expression and use it to develop design guidelines. Two examples demonstrate the simplicity of the design rules.
TL;DR: In this paper, the parameter O in the problem of the Nile is tested using the Sequential Probability Ratio Test (SPRT), which can be used to discriminate between two one-sided hypotheses.
Abstract: The parameter O in the problem of the Nile is tested using the Sequential Probability Ratio Test (SPRT). It is observed that the test can be used to discriminate between two one–sided hypotheses.
TL;DR: In this paper, a sequential t test is presented based on a simple function of the usual Student t statistic, which is shown to possess the martingale property that, conditional on previous observations, the expectation of a future increment is zero.
Abstract: SUMMARY A sequential t test is presented based on a simple function of the usual Student t statistic. This function is shown to possess the martingale property that, conditional on previous observations, the expectation of a future increment is zero. This property facilitates calculation of error probabilities and expected sample sizes, so that the resulting test is relatively straightforward to use. It is argued that this test behaves in a manner similar to that of the locally most powerful test for the mean of a normal distribution with variance known against the one-sided alternative, and might be viewed as its logical analogue for a test of the same hypotheses for the case in which the variance is unknown.
Book•
01 Jan 1989
TL;DR: Reliability distributions multicomponent systems stand-by redundancy life testing the sequential probability ratio test availability and maintainability military standards.
Abstract: Reliability distributions multicomponent systems stand-by redundancy life testing the sequential probability ratio test availability and maintainability military standards.