Spelling suggestions: "subject:"kolmogorovsmirnov test"" "subject:"kolmogorovsmirnov est""
1 
Performance Evaluation of Identification Methods for the Stress Calls of Squirrelfishes¡]Pisces:Holocentridae¡^Tsai, YingWei 25 January 2008 (has links)
In the study of sound identification, land animals such as birds and bats have been well investigated, and so are their habitats. On the other hand, sound making creatures in the ocean are much less researched. In this research, the stress calls of three Holocentridaes, Neoniphon sammara, Myripristis murdjan, and Sargocentron spinosissimum, who are commonly found in coral reefs, were recorded in water tank for analysis of sound characteristics. The averaged characteristic parameters of single pulse among three is around 410 Hz for the peak frequency, 100 Hz for the bandwidth, 0.07 dB/Hz for the slope, and duration of 0.05 s. As for the impulse train, averaged peak frequency is 415 Hz, 55 Hz for the bandwidth, 0.07 dB/Hz for the slope, and duration of 0.5 s. These parameters were first checked by the KolmogorovSmirnov Test to identify if each parameter follows normal distribution; the slopes of ascending and descending frequency and the total duration are not in normal distribution. The three parameters were later transferred so as to concentrate variances. Next, analysis of variance was applied on all characteristics to extract the significant parameters (including non transferred and transferred data), which were then tested by Stepwise Discriminat and Backpropagation Network. The identification rate of for single pulse with and without data transfer is 63% and 82% while pulse train is 57% and 73%. Both identification rates were raised up approximately 20% due to the data transfer. Both methods provide an reliable tool for marine sound identification, and the whole process of the study may be applied to another biological identification.

2 
The Power of Categorical GoodnessOfFit StatisticsSteele, Michael C., n/a January 2003 (has links)
The relative power of goodnessoffit test statistics has long been debated in the literature. ChiSquare type test statistics to determine 'fit' for categorical data are still dominant in the goodnessoffit arena. Empirical Distribution Function type goodnessoffit test statistics are known to be relatively more powerful than ChiSquare type test statistics for restricted types of null and alternative distributions. In many practical applications researchers who use a standard ChiSquare type goodnessoffit test statistic ignore the rank of ordinal classes. This thesis reviews literature in the goodnessoffit field, with major emphasis on categorical goodnessoffit tests. The continued use of an asymptotic distribution to approximate the exact distribution of categorical goodnessoffit test statistics is discouraged. It is unlikely that an asymptotic distribution will produce a more accurate estimation of the exact distribution of a goodnessoffit test statistic than a Monte Carlo approximation with a large number of simulations. Due to their relatively higher powers for restricted types of null and alternative distributions, several authors recommend the use of Empirical Distribution Function test statistics over nominal goodnessoffit test statistics such as Pearson's ChiSquare. Indepth power studies confirm the views of other authors that categorical Empirical Distribution Function type test statistics do not have higher power for some common null and alternative distributions. Because of this, it is not sensible to make a conclusive recommendation to always use an Empirical Distribution Function type test statistic instead of a nominal goodnessoffit test statistic. Traditionally the recommendation to determine 'fit' for multivariate categorical data is to treat categories as nominal, an approach which precludes any gain in power which may accrue from a ranking, should one or more variables be ordinal. The presence of multiple criteria through multivariate data may result in partially ordered categories, some of which have equal ranking. This thesis proposes a modification to the currently available KolmogorovSmirnov test statistics for ordinal and nominal categorical data to account for situations of partially ordered categories. The new test statistic, called the Combined KolmogorovSmirnov, is relatively more powerful than Pearson's ChiSquare and the nominal KolmogorovSmirnov test statistic for some null and alternative distributions. A recommendation is made to use the new test statistic with higher power in situations where some benefit can be achieved by incorporating an Empirical Distribution Function approach, but the data lack a complete natural ordering of categories. The new and established categorical goodnessoffit test statistics are demonstrated in the analysis of categorical data with brief applications as diverse as familiarity of defence programs, the number of recruits produced by the Merlin bird, a demographic problem, and DNA profiling of genotypes. The results from these applications confirm the recommendations associated with specific goodnessoffit test statistics throughout this thesis.

3 
Clusters Identification: Asymmetrical CaseMao, Qian January 2013 (has links)
Cluster analysis is one of the typical tasks in Data Mining, and it groups data objects based only on information found in the data that describes the objects and their relationships. The purpose of this thesis is to verify a modified Kmeans algorithm in asymmetrical cases, which can be regarded as an extension to the research of Vladislav Valkovsky and Mikael Karlsson in Department of Informatics and Media. In this thesis an experiment is designed and implemented to identify clusters with the modified algorithm in asymmetrical cases. In the experiment the developed Java application is based on knowledge established from previous research. The development procedures are also described and input parameters are mentioned along with the analysis. This experiment consists of several test suites, each of which simulates the situation existing in real world, and test results are displayed graphically. The findings mainly emphasize the limitations of the algorithm, and future work for digging more essences of the algorithm is also suggested.

4 
Misbehaving Relay Detection for Cooperative Communications without the Knowledge of Relay MisbehaviorsLi, Chiehkun 17 July 2012 (has links)
In the cooperative communications, the users relay each other's signal and thus form multiple transmission paths to the destination and therefore the system can achieve spatial diversity gain.
Most studies in the literature assumed that cooperative users acting as the relays are normally operated and trustworthy. However, this may not always be true in practice. When the relay misbehaviors are present in the cooperative communications, the communication performance may degrade dramatically and the users may be even better off without cooperation. Therefore, it is necessary for the destination to determine the misbehaving relays and to take appropriate actions
to ensure that cooperative advantages are preserved.
This thesis considers both models in which the cooperative communications are with direct path (WDP) and without direct path (WODP).
Utilizing the proposed KolmogorovSmirnov test mechanism, the destination identifies the misbehaving relays within the cooperative
communications and then excludes their transmitting messages when performing the diversity combining to infer the symbols of interest sent by the source.
In addition, this thesis provides the bit error rate (BER) analysis of the cooperative communications
employing the proposed misbehaving relay detectors. The simulation results demonstrate that the proposed methods have robust performance when the relay misbehaviors are present in the cooperative communications.

5 
New techniques for vibration condition monitoring : Volterra kernel and KolmogorovSmirnovAndrade, Francisco Arruda Raposo January 1999 (has links)
This research presents a complete review of signal processing techniques used, today, in vibration based industrial condition monitoring and diagnostics. It also introduces two novel techniques to this field, namely: the KolmogorovSmirnov test and Volterra series, which have not yet been applied to vibration based condition monitoring. The first technique, the KolmogorovSmirnov test, relies on a statistical comparison of the cumulative probability distribution functions (CDF) from two time series. It must be emphasised that this is not a moment technique, and it uses the whole CDF, in the comparison process. The second tool suggested in this research is the Volterra series. This is a nonlinear signal processing technique, which can be used to model a time series. The parameters of this model are used for condition monitoring applications. Finally, this work also presents a comprehensive comparative study between these new methods and the existing techniques. This study is based on results from numerical and experimental applications of each technique here discussed. The concluding remarks include suggestions on how the novel techniques proposed here can be improved.

6 
Automatic measurement of particles from holograms taken in the combustion chamber of a rocket motorCarrier, Denis Joseph Gaston 12 1900 (has links)
Approved for public release; distribution is unlimited / This thesis describes the procedure used for the automatic measurement of particles from hologram taken in
the combustion chamber of a rocket motor while firing. It describes the investigation done on two averaging
techniques used to reduce speckle noise, capturing the image focused on a spinning mylar disk and software
averaging of several image frames. The spinning disk technique proved superior for this application. The
KolmogorovSmirnov twosample test is applied to
different particle samples in order to find an estimate of
the number of particles required to obtain a stable
distribution function. The number of particles is
calculated and given. The last part of this study shows
real particle distributions in the form of frequency
histograms. / http://archive.org/details/automaticmeasure00carr / Major, Canadian Armed Forces

7 
A KolmogorovSmirnov Test for r SamplesBöhm, Walter, Hornik, Kurt 12 1900 (has links) (PDF)
We consider the problem of testing whether r (>=2) samples are drawn from the same continuous distribution F(x). The test statistic we will study in some detail is defined as the maximum of the circular differences of the empirical distribution functions, a generalization of the classical 2sample KolmogorovSmirnov test to r (>=2) independent samples. For the case of equal sample sizes we derive
the exact null distribution by counting lattice paths confined to stay in the scaled alcove $\mathcal{A}_r$ of the affine Weyl group $A_{r1}$. This is done using a generalization of the classical reflection principle. By a standard diffusion scaling we derive also the asymptotic distribution of the test statistic in terms of a multivariate Dirichlet series. When the sample sizes are not equal the reflection principle no longer works, but we are able to establish a weak convergence result even
in this case showing that by a proper rescaling a test statistic based on a linear transformation of the circular differences of the empirical distribution functions has the
same asymptotic distribution as the test statistic in the case of equal sample sizes. / Series: Research Report Series / Department of Statistics and Mathematics

8 
The Impact of Midbrain Cauterize Size on Auditory and Visual Responses' DistributionZhang, Yan 20 April 2009 (has links)
This thesis presents several statistical analysis on a cooperative project with Dr. Pallas and Yuting Mao from Biology Department of Georgia State University. This research concludes the impact of cauterize size of animals’ midbrain on auditory and visual response in brains. Besides some already commonly used statistical analysis method, such as MANOVA and Frequency Test, a unique combination of Permutation Test, KolmogorovSmirnov Test and Wilcoxon Rank Sum Test is applied to our nonparametric data. Some simulation results show the Permutation Test we used has very good powers, and fits the need for this study. The result confirms part of the Biology Department’s hypothesis statistically and enhances more complete understanding of the experiments and the potential impact of helping patients with Acquired Brain Injury.

9 
Deviating timetoonset in predictive models : detecting new adverse effects from medicinesWärn, Caroline January 2015 (has links)
Identifying previously unknown adverse drug reactions becomes more important as the number of drugs and the extent of their use increases. The aim of this Master’s thesis project was to evaluate the performance of a novel approach for highlighting potential adverse drug reactions, also known as signal detection. The approach was based on deviating timetoonset patterns and was implemented as a twosample KolmogorovSmirnov test for nonvaccine data in the safety report database, VigiBase. The method was outperformed by both disproportionality analysis and the multivariate predictive model vigiRank. Performance estimates indicate that deviating timetoonset patterns is not a suitable approach for signal detection for nonvaccine data in VigiBase.

10 
Possible Difficulties in Evaluating University PerformanceBased on Publications Due to Power Law Distributions : Evidence from SwedenSadric, Haroon, Zia, Sarah January 2023 (has links)
Measuring the research performance of a university is important to the universities themselves, governments, and students alike. Among other metrics, the number of publications is easy to obtain, and due to the large number of publications each university produces during one year, it suggests to be one accurate metric. However, the number of publications depends largely on the size of the institution, suggesting, if not addressed, that larger universities are better. Thus, one might intuitively try to normalize by size and use publications per researcher instead. A better institution would allow individual researchers to have more publications each year. However, publications, like many other things, might follow a powerlaw distribution, where most researchers have few, and only a few researchers have very many publications. These powerlaw distributions violate the assumptions the central limit the orem makes, for example, having a welldefined mean or variance. Specifically, one can not normalize or use averages from powerlaw distributed data, making the comparison of university publications impossible if they indeed follow a powerlaw distribution. While it has been shown that some scientific domains or universities show this powerlaw distribution, it is not known if Swedish universities also show this phenomenon. Thus, here we collect publication data for Swedish universities and determine whether or not, they are powerlaw distributed. Interestingly, if they are, one might use the slope of the powerlaw distribution as a proxy to determine research output. If the slope is steep, it suggests that the ratio between highly published authors and those with few publications is small. Where as a flatter slope suggests that a university has more highly published authors than a university with a steeper slope. Thus, the second objective here is to assess if the slope of the distribution can be determined or to which extent this is possible. This study will show that eight of the fifteen Swedish universities considered follow a powerlaw distribution (KolmogorovSmirnov statistic<0.05), while the remaining seven do not. The key determinant is the total number of publications. The difficulty here is that often the total number of publications is so small that one can not reject a powerlaw distribution, and it is also impossible to determine the slope of the distribution with any accuracy in those cases. While this study suggests that in principle, the slopes of the powerlaw distributions can be used as a comparative metric, it also showed that for half of Sweden’s universities, the data is insufficient for this type of analysis.

Page generated in 0.0451 seconds