scispace - formally typeset
Search or ask a question
Journal ArticleDOI

Nonparametric Partial Sequential Test for Location Shift at an Unknown Time Point

22 Jan 2007-Sequential Analysis (Taylor & Francis Group)-Vol. 26, Iss: 1, pp 99-113
TL;DR: In this paper, a partial sequential sampling scheme is introduced to develop a sequential rank-based nonparametric test for the identity of two unknown univariate continuous distribution functions against one-sided shift in location occurring at an unknown time point.
Abstract: In the present work, we introduce a partial sequential sampling scheme to develop a sequential rank-based nonparametric test for the identity of two unknown univariate continuous distribution functions against one-sided shift in location occurring at an unknown time point. Our work is motivated by Wolfe (1977) as well as Orban and Wolfe (1980). We provide detailed discussion on asymptotic studies related to the proposed test. We compare the proposed test with a usual rank-based test. Some simulation studies are also presented.
Citations
More filters
Journal ArticleDOI
TL;DR: A new nonparametric control chart based on the Ansari–Bradley non parametric test and the effective change point model is developed and simulation results show that the proposed control chart is superior to other non Parametric control charts in monitoring process variability for most cases.
Abstract: Statistical process control is widely used in industrial processes, service fields, among others. While parametric control charts are useful in certain processes, there is often a lack of enough knowledge about the process distribution. So, nonparametric control charts are needed in such situations. This paper develops a new nonparametric control chart based on the Ansari–Bradley nonparametric test and the effective change point model. Simulation results show that our proposed control chart is superior to other nonparametric control charts in monitoring process variability for most cases. Our proposed control chart is easy in computation, and powerful for monitoring process variability. Copyright © 2016 John Wiley & Sons, Ltd.

24 citations

Journal ArticleDOI
TL;DR: In this paper, partial sequential nonparametric tests for multiple comparison are presented. Butler et al. provide tests for the identity of several unknown univariate continuous distribution functions against patterned alternatives based on Wilcoxon score.
Abstract: In the present work we develop partial sequential nonparametric tests for multiple comparison. We provide tests for the identity of several unknown univariate continuous distribution functions against patterned alternatives. Our tests are based on Wilcoxon score. We conduct some Monte Carlo studies related to the proposed tests. We carry out a detailed comparison between the proposed procedures and the corresponding nonsequential procedures. We register significant gain in sample size through the proposed procedure, maintaining almost the same level and power for both the tests. We perform an analysis of life data arising out of a geological survey related to arsenic contamination. We also present some asymptotics in this context.

15 citations


Cites background from "Nonparametric Partial Sequential Te..."

  • ...We assume that, for each m there are positive numbers ri = ri m and positive integers i = i m i = 1 2 s such that, as m → , i → but ri m i m → i ∈ 0 (4.1) Set T imk = 1√ ri ( Fm Yik − 1 2 ) k = 1 2 i i = 1 2 s Then, as in Bandyopadhyay and Mukherjee (2007), we observe the following simple results....

    [...]

  • ...Then, as in Bandyopadhyay and Mukherjee (2007), we observe the following simple results....

    [...]

  • ...In fact, as in Bandyopadhyay and Mukherjee (2007), it can be seen that ami i = 1 2 s converges to an s-variate normal distribution with mean vector 0 and...

    [...]

  • ...…null distribution of ∑ i k=1 T i mk is normal with mean ami and variance v 2 m where ami = im 1√ri ∑m k=1 1 2 − F Xi and v2m = E Fm Yk − 12 2 In fact, as in Bandyopadhyay and Mukherjee (2007), it can be seen that ami i = 1 2 s converges to an s-variate normal distribution with mean vector 0…...

    [...]

Journal ArticleDOI
TL;DR: Some asymptotically power is developed on partially sequential nonparametric tests for monitoring structural changes based on Wilcoxon score and it is seen that one of the proposed procedures significantly controls the Type I error rate.
Abstract: In the present article, we develop some asymptotically power on partially sequential nonparametric tests for monitoring structural changes. Our test procedures are based on Wilcoxon score. We use the idea of curved stopping boundaries. We derive some exact results and perform simulation studies to provide various properties of the tests. We see that one of the proposed procedures significantly controls the Type I error rate. This procedure may be very effective for fluctuation monitoring. We illustrate the procedures by using real life data from the stock market.

12 citations


Cites background or methods from "Nonparametric Partial Sequential Te..."

  • ...Recently, Bandyopadhyay and Mukherjee (2007) updated the partially sequential stopping rule using the concept of sequential ranks. Such sequential ranks were used earlier in connection with a quasi-sequential stopping rule by Bhattacharya and Frierson (1981). However, all such sequential stopping rules terminate with probability one under both the null and alternative hypotheses....

    [...]

  • ...Recently, Bandyopadhyay and Mukherjee (2007) updated the partially sequential stopping rule using the concept of sequential ranks. Such sequential ranks were used earlier in connection with a quasi-sequential stopping rule by Bhattacharya and Frierson (1981). However, all such sequential stopping rules terminate with probability one under both the null and alternative hypotheses. Thus, the monitoring stops even when there is no fluctuation in the populations. This is really unwarranted in much econometric as well as environmental monitoring. Here a process needs to be monitored ceaselessly. Particularly, when we assume a negligible cost of sampling under no fluctuation, we do not need to stop at all unless there is a signal. Chu et al. (1996) emphasizes the need for developing partial sequential tests which rarely terminate when no fluctuation is observed....

    [...]

  • ...Recently, Bandyopadhyay and Mukherjee (2007) updated the partially sequential stopping rule using the concept of sequential ranks. Such sequential ranks were used earlier in connection with a quasi-sequential stopping rule by Bhattacharya and Frierson (1981). However, all such sequential stopping rules terminate with probability one under both the null and alternative hypotheses. Thus, the monitoring stops even when there is no fluctuation in the populations. This is really unwarranted in much econometric as well as environmental monitoring. Here a process needs to be monitored ceaselessly. Particularly, when we assume a negligible cost of sampling under no fluctuation, we do not need to stop at all unless there is a signal. Chu et al. (1996) emphasizes the need for developing partial sequential tests which rarely terminate when no fluctuation is observed. They also discussed the importance of controlling Type I error in monitoring structural changes. Thus, we require framing a rule in such a way that a termination will signal instability. Hence we certainly need to construct tests with asymptotically or approximately power one and simultaneously to achieve control over Type I error rate. Sequential procedures are effective in several real bio-statistical and econometric problems. Sequential plans often save precious sample sizes and give efficient inference. On the other hand, in most of the real life situations, the normality assumption does not work well. It is difficult to suggest suitable non normal model in many situations. Moreover, inference based on non normal models is not always easy. Thus, nonparametric models have a wide applicability in the present era. At the same time, research related to controlling Type I error rate in statistical inference are getting more and more importance. As the pattern of the alternative is unknown or vague in most cases, deriving optimal tests become complicated. Statisticians are, in these days, avoiding the long practice of using mathematically sound optimal tests. Those tests, mainly in the area of sequential clinical trials, are continually being replaced by some heuristic tests that are more practical and easy to handle. If properly designed, such heuristic tests are likely to have greater appeal in testing some econometric or environmental hypothesis as well. One may see Huang et al. (2005) for an illustration of controlling Type I error in adjusted O’Brien’s test....

    [...]

  • ...Bandyopadhyay and Mukherjee (2007) show for linear stopping boundary that sequential rank test can even improve power in such occasion if a shift occurs at a later stage....

    [...]

  • ...…by the following stopping variable: M = min { n n∑ k=1 Fm Yk ≥ r/2 } (2.1) where r is a prefixed positive number and Fm · is the empirical df based on Xm. Bandyopadhyay and Mukherjee (2007) suggest a stopping rule by updating the empirical df using the available second stage observations also....

    [...]

Journal ArticleDOI
TL;DR: In this article, the authors address the issues identifying warning and action points in a time sequential monitoring of the exchange rate records and develop some approximate power 1 partially sequential nonparametric tests that are based on both usual and sequential ranks.
Abstract: Financial or economic crisis of 2008 and its severe negative impact on the human race is currently one of the focal themes of financial research. During this crisis period the Indian Rupee (INR) has significantly lost its ground compared to the United States Dollar (USD). In the present paper, we address the issues identifying warning and action points in a time sequential monitoring of the exchange rate records. We developed some approximate power 1 partially sequential nonparametric tests that are based on both usual and sequential ranks. Proposed procedures necessarily help us to capitalize the best features of the two different rank procedures. We use Wilcoxon score and the idea of curved stopping boundaries as in Bandyopadhyay et al. (2008). We vividly present our numerical findings obtained by sequential Monte-Carlo procedures. We aim at controlling the type I error rate in course of structural monitoring.

11 citations


Cites background or methods from "Nonparametric Partial Sequential Te..."

  • ...and sequential rank-based Bandyopadhyay and Mukherjee (2007)-type stopping rule becomes...

    [...]

  • ...We may readily see from Bandyopadhyay and Mukherjee (2007) that, under H0, both Tn and Sn have symmetric distributions around the common expected value n/2, although Tn has larger variance than Sn....

    [...]

  • ...Among others, to name a few are Page (1955), Bhattacharya and Johnson (1968), and Bandyopadhyay and Mukherjee (2007). As in Bandyopadhyay and Mukherjee (2007), we define Fm · be the empirical d....

    [...]

  • ...4 of Bandyopadhyay and Mukherjee (2007) and can easily be extended for bivariate situation employing Cramer–Wold device....

    [...]

  • ...Among others, to name a few are Page (1955), Bhattacharya and Johnson (1968), and Bandyopadhyay and Mukherjee (2007). As in Bandyopadhyay and Mukherjee (2007), we define Fm · be the empirical d.f. based on m. Further suppose Hm+k−1 · be the empirical distribution function (d.f.) based on m and 1 2 k−1 , k ≥ 1. Then Tn = ∑n k=1 Fm k gives the usual rank sum statistic, whereas the sequential rank sum statistic is given by Sn = ∑n k=1 Hm+k−1 k . Hence usual rank-based Orban and Wolfe (1980)-type stopping rule becomes...

    [...]

Journal ArticleDOI
TL;DR: In this paper, a partially sequential sampling procedure was proposed to develop a nonparametric method for simultaneous testing, which is motivated by an interesting investigation related to arsenic contamination in ground water.
Abstract: In the present paper we introduce a partially sequential sampling procedure to develop a nonparametric method for simultaneous testing. Our work, as in [U. Bandyopadhyay, A. Mukherjee, B. Purkait, Nonparametric partial sequential tests for patterned alternatives in multi-sample problems, Sequential Analysis 26 (4) (2007) 443–466], is motivated by an interesting investigation related to arsenic contamination in ground water. Here we incorporate the idea of multiple hypotheses testing as in [Y. Benjamini, T. Hochberg, Controlling the false discovery rate: A practical and powerful approach to multiple testing, Journal of Royal Statistical Society B 85 (1995) 289–300] in a typical way. We present some Monte Carlo studies related to the proposed procedure. We observe that the proposed sampling design minimizes the expected sample sizes in different situations. The procedure as a whole effectively describes the testing under dual pattern alternatives. We indicate in brief some large sample situations. We also present detailed analysis of a geological field survey data.

11 citations

References
More filters
Book
01 Jan 1967
TL;DR: In this article, the authors present an elementary theory of rank tests and a set of properties of rank estimators, including asymptotic optimality and efficiency, as well as non-null distributions.
Abstract: Introduction and Coverage. Preliminaries. Elementary Theory of Rank Tests. Selected Rank Tests. Computation of Null Exact Distributions. Limiting Null Distributions. Limiting Non-Null Distributions. Asymptotic Optimality and Efficiency. Rank Estimates and Asymptotic Linearity.

1,653 citations

Book
01 Jan 1979
TL;DR: In this paper, the distribution-free statistics under the null hypothesis are presented. But they do not consider the one-sample location problem, and the scale problem is not considered.
Abstract: Distribution-Free Statistics. Power Functions and Their Properties. Asymptotic Relative Efficiency of Tests. Confidence Intervals and Bounds. Point Estimation. Linear Rank Statistics Under the Null Hypothesis. Two-Sample Location and Scale Problems. The One-Sample Location Problem. Additional Methods for Constructing Distribution-Free Procedures. Other Important Problems. Appendix. Index.

758 citations

Journal ArticleDOI
E. S. Page1

674 citations


"Nonparametric Partial Sequential Te..." refers background in this paper

  • ...Testing a problem of this type related to the location shift at an unknown time point was first introduced by Page (1955) for a known initial level....

    [...]

Book
01 Jan 1981

180 citations


"Nonparametric Partial Sequential Te..." refers background in this paper

  • ...Our work is motivated by Wolfe (1977) as well as Orban and Wolfe (1980)....

    [...]

  • ...For various inferential problems based on usual ranks, one can also go through the book by Sen (1981)....

    [...]