Search

This paper used different copula-based GARCH models (Copula-GARCH model and Copula-GJR-GARCH model) to analyze the dependence structure among gold price, stock price index of gold mining companies and Shanghai Composite Index in China. The empirical results found that the suitable margins were skew-t distribution, and the GJR-GARCH marginal distribution had better explanatory ability than the GARCH model. Moreover, we found the Clayton copula had the highest explanatory ability of the dependence structure for all series. There existed positive dependence in the rates of return for all series, and the dependence between these markets will be closer with the gradual integration of international financial market.

Within the genus Rosa numerous species have been described. Circumscription of the dogrose section Caninae is straightforward, but the delineation of species and subsections within this section is less clear, partly due to hybridisation between species. We have investigated the extent to which DNA marker-based information of wild populations corroborates present-day dogrose taxonomy and hypotheses about the origination of taxa. Sampling was conducted in a transect across Europe, collecting over 900 specimens of all encountered dogrose taxa. For comparison, we also included more than 200 samples of species belonging to other sections. Two lines of statistical analyses were used to investigate the genetic structure based on AFLP data: (1) an unstructured model with principal coordinate analysis and hierarchical clustering, and (2) a model with a superimposed taxonomic structure based on analysis of genetic diversity using a novel approach combining assignment tests with canonical discriminant analysis. Support was found for five of the seven subsections, whereas R. balsamica apparently belongs to subsection Caninae thus omitting the need for recognising subsection Tomentellae. For R. stylosa, a hybridogenic origin with a non-dogrose section member has been suggested, and it can be treated either as a separate subsection or within subsection Caninae. Within the subsection Rubigineae, a species cluster with low support for the taxa R. micrantha, R. rubiginosa and the putatively hybridogenous R. gremlii was identified. Similarly, several species in the subsection Caninae overlapped considerably, and are best regarded as one common species complex. This population genetic approach provides a general method to validate the taxonomic system in complex and polyploid taxa.

Agreement among raters is an important issue in medicine, as well as in education and psychology. The agreement among two raters on a nominal or ordinal rating scale has been investigated in many articles. The multi-rater case with normally distributed ratings has also been explored at length. However, there is a lack of research on multiple raters using an ordinal rating scale. In this simulation study, several methods were compared with analyze rater agreement. The special case that was focused on was the multi-rater case using a bounded ordinal rating scale. The proposed methods for agreement were compared within different settings. Three main ordinal data simulation settings were used (normal, skewed and shifted data). In addition, the proposed methods were applied to a real data set from dermatology. The simulation results showed that the Kendall's W and mean gamma highly overestimated the agreement in data sets with shifts in data. ICC4 for bounded data should be avoided in agreement studies with rating scales

Consider a centralized query operator that identifies to every smart phone user its k geographically nearest neighbors at all times, a query we coin Continuous All k-Nearest Neighbor (CAkNN). Such an operator could be utilized to enhance public emergency services, allowing users to send SOS beacons out to the closest rescuers and allowing gamers or social networking users to establish ad-hoc overlay communication infrastructures, in order to carry out complex interactions. In this paper, we study the problem of efficiently processing a CAkNN query in a cellular or WiFi network, both of which are ubiquitous. We introduce an algorithm, coined Proximity, which answers CAkNN queries in O(n(k+λ)) time, where n denotes the number of users and λ a network-specific parameter (λ ≪ n). Proximity does not require any additional infrastructure or specialized hardware and its efficiency is mainly attributed to a smart search space sharing technique we introduce. Its implementation is based on a novel data structure, coined k+-heap, which achieves constant O(1) look-up time and logarithmic O(log(k*λ .)) insertion/update time. Proximity, being parameter-free, performs efficiently in the face of high mobility and skewed distribution of users (e.g., the service works equally well in downtown, suburban, or rural areas). We have evaluated Proximity using mobility traces from two sources and concluded that our approach performs at least one order of magnitude faster than adapted existing work. © 2012 IEEE. 79 88 Sponsors: IBM MICS Indian Institute of Technology Microsoft Research Bell Labs Conference code: 94301 Cited By :18

Regression trees represent one of the most popular tools in predictive data mining applications. However, previous studies have shown that their performances are not completely satisfactory when the dependent variable is highly skewed, and severely degrade in the presence of heavy-tailed error distributions, especially for grossly mis-measured values of the dependent variable. In this paper the lack of robustness of some classical regression trees is investigated by addressing the issue of highly-skewed and contaminated error distributions. In particular, the performances of some non robust regression trees are evaluated through a Monte Carlo experiment and compared to those of some trees, based on M-estimators, recently proposed in order to robustify this kind of methods. In conclusion, the results obtained from the analysis of a real dataset are presented.

Regression trees represent one of the most popular tools in predictive data mining applications. However, their performances severely degrade in the presence of highly-skewed and/or long-tailed error distributions, and especially for grossly mis-measured values of the dependent variable. In this paper, these issues are discussed from both a theoretical and a practical point of view, and some recent proposals to overcome these difficulties are presented.

I study the incentives of Cournot duopolists to share their technologies with their competitor in markets where intellectual property rights are absent and imitation is costless. The trade-off between a signaling effect and an expropriation effect determines the technology-sharing incentives. In equilibrium at most one firm shares some of its technologies. For similar technology distributions, there exists an equilibrium in which nobody shares. If the technology distributions are skewed towards efficient technologies, then there may exist equilibria in which one firm shares all technologies, only the best technologies, or only intermediate technologies. No other equilibria can exist.

Question Is native species occurrence related to soil nutrients in highly invaded Californian annual grasslands? What is the best method to analyze this relationship, given that native species occur in very low numbers and are absent from many locations? Location California, USA. Methods We investigated the effects of soil characteristics and livestock grazing on native plant occurrence at 40 plots from six sites during the period 2003-2005. Low absolute cover (< 5.8%) of native species resulted in strongly skewed, zero-inflated data sets. To overcome problems in the analysis created by non-normality and correlations within plots, we used GLMs and GLMMs, either with a Poisson or a negative binomial distribution, to analyse native species richness and Nassella pulchra cover. Results N. pulchra cover was strongly associated with low phosphorus in sandy soils, whereas native species richness was highest in soils with low available nitrogen (high CN). Conclusion Under current conditions, phosphorus seems to be a critical factor influencing abundance of N. pulchra. Low fertility soils may provide refugia for native species in highly invaded California grasslands as they may be below a threshold required for non-native annuals to completely dominate. By using non-normal distributions in linear models with random components, we report well fitted models with more accurately tested significant covariates. © IAVS; Opulus Press.

An allopolyploid complex with high genomic integrity has been studied. Dogroses transmit only seven chromosomes (from seven bivalents) through the pollen, whereas 21, 28 or 35 chromosomes (from seven bivalents and 14, 21 or 28 univalents) come from the egg cells. Seedlings derived from two interspecific crosses were analysed with flow cytometry and molecular markers to determine ploidy level, mode of reproduction and genomic constitution. Evidence was obtained for the formation of unreduced male and female gametes, which can take part in fertilization (producing seedlings with higher ploidy than the parental plants) or in apomictic reproduction. Random amplified polymorphic DNA (RAPD) and microsatellite analyses indicated that three seedlings (5%) were derived through apomixis, whereas the other 49 were hybrids. Bivalent formation appears to involve chromosomes that consistently share the same microsatellite alleles. Allele-sharing between the maternally transmitted and highly conserved univalent-forming chromosomes reflected the taxonomic distance between different genotypes. The frequently recombining bivalent-forming chromosomes were taxonomically less informative.

According to previous cytological evidence, the hemisexual dog-rose species, Rosa sect. Caninae, transmit only seven chromosomes (derived from seven bivalents) through their pollen grains, whereas egg cells contain 21, 28 or 35 chromosomes (derived from seven bivalents and 14, 21 or 28 univalents) depending on ploidy level. Two sets of reciprocal pairwise interspecific crosses involving the pentaploid species pair R. dumalis and R. rubiginosa, and the pentaploid/tetraploid species pair R. sherardii and R. villosa, were analysed for 13 and 12 microsatellite DNA loci, respectively. Single loci were represented by a maximum of three simultaneously occurring alleles in R. villosa, and four alleles in the other three parental plants. In the experimentally derived offspring, the theoretical maximum of five alleles was found for only one locus in the pentaploid progenies. Microsatellite DNA allele composition was identical with that of the maternal parent in 10 offspring plants, which were probably derived through apomixis. Almost all microsatellite DNA alleles were shared with the maternal parent also in the remaining offspring, but 1-4 alleles shared only with the paternal parent, indicating sexual seed formation. Analysis of quantitative peak differences allowed a tentative estimation of allelic configuration in the individual plants, and suggested that bivalent formation preferentially takes place between chromosomes that consistently share the same microsatellite alleles and therefore appear to be highly homologous. Moreover, alleles that were shared between the species in each cross combination comparatively often appear to reside on the bivalent-forming chromosomes, whereas species-specific alleles instead occur comparatively often on the univalent-forming chromosomes and are therefore inherited through the maternal parent only. Recombination then takes place between very similar genomes also in interspecific crosses, resulting in a reproduction system that is essentially a mixture between apomixis and selfing.

In this letter, we introduce a novel family of texture models which provide alternatives to texture models which are based on Gaussian distributions. In particular, we introduce linear textures generated with a member of the alpha-stable distribution family which is a generalisation of the Gaussian distribution. The new family of texture models is capable of representing both impulsive and unsymmetric (skewed) image data which cannot be accommodated by the Gaussian model. We present new techniques for texture model estimation and we demonstrate the success of the techniques on synthetic data.

Mining applications commonly faces surprising high values designated as outliers. These values impact dramatically statistical analysis and interpretation. A comprehensive analysis on the causes for the presence of unexpected high values was recommended. However, if an erroneous value was accepted as a part of the solution, some form of correction is recommended. A methodology based on the robust kriging (RoK) algorithm is proposed to be used in exploratory data analysis and also to deal with problems associated with the presence of outliers in the sample data set. The efficiency of RoK method as an interpolator is tested in different types of mineralizations. Importantly, the parent population from which the data was sampled is available, thus allowing direct quantitative assessment of the effectiveness of the estimation technique. The performance of the method is tested in the context of ore reserves estimation. RoK model is compared to models generated by ordinary kriging, median indicator kriging, and lognormal kriging. RoK proved to be more accurate and more precise than those methods reducing substantially the number of misclassified blocks

Bu makelede, genelde analitik bir ifadeye sahip olmayan pozitif alpha-kararli olasilik yogunluk fonksiyonu icin analitik bir yaklastirma 'oneriyoruz. 'Onerdigimiz yaklastirma, pozitif alpha-kararli rastsal degiskenlerin Pearson ve diger bir pozitif alpha-kararli dagilima c'oz'umlenmesine dayanmakta. Bu c'oz'umleme herhangi bir pozitif alpha-kararli dagilimin birer Pearson karisimi seklinde ifade edilebilmesini saglamakta.

In this letter, we introduce a novel family of texture models which provide alternatives to texture models which are based on Gaussian distributions. In particular, we introduce linear textures generated with a member of the a-stable distribution family which is a generalisation of the Gaussian distribution. The new family of texture models is capable of representing both impulsive and unsymmetric (skewed) image data which cannot be accommodated by the Gaussian model. We present new techniques for texture model estimation and we demonstrate the success of the techniques on synthetic data.

High anomalous grades are very common in gold deposits, whose presence requires careful treatment to prevent overestimation of metal content. Mineral resource analysts have worked on the estimation of several gold deposits, and none of the classical methods were able to avoid manual interventions, such as cutting high grades for local estimation or using more information beyond the data for the variogram inference. The Field Parametric Geostatistics (FPG) is presented as an alternative for the application of linear kriging methods to estimate highly skewed distributions, proposing a mathematical model which incorporates the grades and its representativeness into a new variable, reducing the influence of high grades without empirical manual interventions. In this article, the mathematical formulation of the FPG theory is presented, as well as its application in datasets with outliers and high skewed distributions: the Walker Lake dataset and the Amapari gold deposit. The results are compared to results obtained by the application of standard techniques, demonstrating that FPG is a feasible alternative to estimate local grades and local reserves for highly skewed variables.