Regression models with correlated errors lead to nonadditivity of the information matrix. This makes the usual approach of design optimization (approximation with a continuous design, application of an equivalence theorem, numerical calculations by a gradient algorithm) impossible. A method is presented that allows the construction of a gradient algorithm by altering the information matrices through adding of supplementary noise. A heuristic is formulated to circumvent the nonconvexity problem and the method is applied to typical examples from the literature. (author's abstract) / Series: Forschungsberichte / Institut für Statistik

Hedging a contingent claim with an asset which is not perfectly correlated with the underlying asset results in an imperfect hedge. The residual risk from hedging with a correlated asset is priced using an actuarial standard deviation principle in infinitesmal time, which gives rise to a nonlinear partial differential equation (PDE). A fully implicit, monotone discretization method is developed for solving the pricing PDE. This method is shown to converge to the viscosity solution. Certain grid conditions are required to guarantee monotonicity. An algorithm is derived which, given an initial grid, inserts a finite number of nodes in the grid to ensure that the monotonicity condition is satisfied. At each timestep, the nonlinear discretized algebraic equations are solved using an iterative algorithm, which is shown to be globally convergent. Monte Carlo hedging examples are given, which show the standard deviation of the profit and loss at the expiry of the option.

An achievable rate region for general wireless sensor networks is proposed. A general multi-source, multi-relay, multi-destination wireless sensor network with arbitrarily correlated sources is considered. Each node can sense some real phenomena and send its readings to one or more sinks (data gathering nodes) via some relays. It also can relay some correlated or independent readings of other nodes, simultaneously. In this problem the source and channel coding separation is not optimal and the information which each reading has about others nodes is destroyed in separation. Thus, a joint source channel coding scheme can be used. The problem consists of relay channels and multiple access channels with arbitrarily correlated sources. The proposed scheme is based on regular block Markov encoding/backward decoding and code division multiple-access (CDMA) and the result is a combination of multi-relay and multiple-access with correlated sources.

Classification, an important problem of data mining, is one of useful techniques for prediction. The goal of the classification problem is to construct a classifier from a given database for training, and to predict new data with the unknown class. Classification has been widely applied to many areas, such as medical diagnosis and weather prediction. The decision tree is the most popular model among classifiers, since it can generate understandable rules and perform classification without requiring any computation. However, a major drawback of the decision tree model is that it only examines a single attribute at a time. In the real world, attributes in some databases are dependent on each other. Thus, we may improve the accuracy of the decision tree by discovering the correlation between attributes. The CAM method applies the method of mining association rules, like the Apriori method, for discovering the attribute dependence. However, traditional methods for mining association rules are inefficient in the classification applications and could have five problems: (1) the combinatorial explosion problem, (2) invalid candidates, (3) unsuitable minimal support, (4) the ignored meaningful class values, and (5) itemsets without class data. The FP-growth avoids the first two problems. However, it is still suffered from the remaining three problems. Moreover, one more problem occurs: Unnecessary nodes for the classification problem which make the FP-tree incompact and huge. Furthermore, the workload of the CAM method is expensive due to too many times of database scanning, and the attribute combination problem causes some misclassification. Therefore, in this thesis, we present an efficient and accurate decision tree building method which resolves the above six problems and reduces the overhead of database scanning in the CAM method. We build a structure named class-rooted FP-tree which is a tree similar to the FP-tree, except the root of the tree is always a class item. Instead of using a static minimal support applied in the FP-growth method, we decide the minimal support dynamically, which can avoid some misjudgement of large itemsets used for the classification problem. In the decision tree building phase, we provide a pruning strategy that can reduce the times of database scanning. We also solve the attribute combination problem in the CAM method and improve the accuracy. From our simulation, we show that the performance of the proposed class-rooted FP-tree mining method is better than that of other mining association rule methods in terms of storage usage. Our simulation also shows the performance improvement of our method in terms of the times of database scanning and classification accuracy as compared with the CAM method. Therefore, the mining strategy of our proposed method is applicable to any method for building decision tree, and provides high accuracy in the real world.

In this thesis, we propose and analyze a multicast key backbone for secure group communications. We also utilize the correlated relationships between the mobile users in the wireless communications networks. When a batch member joins or leaves the group communications, the system has to update and distribute encryption keys to assure that only active members could receive the latest information. In previous tree-based multicast key management schemes, the depth of the key tree is unbounded and analytically deriving the exact value of the corresponding average update cost remains an open problem. And in previous schemes, the different mobile user arrives in and leaves from the system at different time. In contrast, the depth of the proposed multicast key backbone is fixed and the arriving or leaving users are more than one. We utilize these two characteristics and simulate the system to get the average update cost per time unit. We can find that this scheme can improve the efficiency of the system in some special cases when updating the new key.

Available fragile site identification software packages (FSM and FSM3) assume that all chromosomal breaks occur independently. However, under a Mendelian model of inheritance, homozygosity at fragile loci implies pairwise correlation between homologous sites. We construct correlation models for chromosomal breakage data in situations where either partitioned break count totals (per-site single-break and doublebreak totals) are known or only overall break count totals are known. We derive a likelihood ratio test and Neyman’s C( α) test for correlation between homologs when partitioned break count totals are known and outline a likelihood ratio test for correlation using only break count totals. Our simulation studies indicate that the C( α) test using partitioned break count totals outperforms the other two tests for correlation in terms of both power and level. These studies further suggest that the power for detecting correlation is low when only break count totals are reported. Results of the C( α) test for correlation applied to chromosomal breakage data from 14 human subjects indicate that detection of correlation between homologous fragile sites is problematic due to sparseness of breakage data. Simulation studies of the FSM and FSM3 algorithms using parameter values typical for fragile site data demonstrate that neither algorithm is significantly affected by fragile site correlation. Comparison of simulated fragile site misclassification rates in the presence of zero-breakage data supports previous studies (Olmsted 1999) that suggested FSM has lower false-negative rates and FSM3 has lower false-positive rates.

ITC/USA 2014 Conference Proceedings / The Fiftieth Annual International Telemetering Conference and Technical Exhibition / October 20-23, 2014 / Town and Country Resort & Convention Center, San Diego, CA / Spatial modulation (SM) is a transmission scheme where only one transmit antenna is active at any time instant. It thus reduces interchannel interference (ICI) and receiver complexity over traditional multi-antenna systems. However, the spectral efficiency of SM is low. To improve the spectral efficiency, generalized spatial modulation (GSM) can be used. In this paper, we propose to apply the Alamouti technique with GSM for correlated antennas, and show that the proposed approach provides significant improvement over conventional SM and GSM. Our study also shows the importance of bit-to-antenna mappings and their roles on the selection of appropriate correlated antennas.

Generalized estimating equations (GEEs) are commonly used in the modelling of correlated data. However, in the presence of informative cluster sizes, estimates obtained using GEEs may be biased. In order to correct for this bias a weighted GEE may be used. Previous research has extended the use of weighted GEEs to a longitudinal setting but requires that cluster sizes remain constant over time. In this thesis, two new weighting schemes are investigated to allow for valid parameter estimation in a longitudinal setting where cluster sizes are informative and may change over time. Specifically, this thesis considers weighting by the inverse of the time-dependent cluster size, and by the total number of observations for a given cluster. Through Monte Carlo simulation, the performance of traditional GEEs, GEEs under previously proposed weighting schemes, and these two new models are compared. Results of these studies show that weighting by the total number of observations results in unbiased parameter estimates with excellent coverage.

One of the principle factors that limits the areas to which seeded bermudagrass can be adapted is low temperature. Therefore, increasing the winter tolerance of seeded bermudagrass cultivars has been a goal of turfgrass breeders for many years. Design of an efficient breeding method for developing cultivars with increased tolerance to cool and freezing temperatures could be enhanced by having heritability estimates for cold tolerance traits. Additionally, the identification of correlated traits can be useful in improvement of cultivar development. Heritability estimates for winter tolerance can be obtained from cold treatments imposed artificially or from observations made in the field. Parental clones and their respective polycross half-sib families were established in a randomized complete block design with four replications in 2004 in Lexington, KY. Differences in spring green up and fall dormancy measured in 2006 were detected between genotypes within the progeny and parental groups. These same lines were subjected to 15 replications of a freezing treatment in a freeze chamber. Differences in spring green up and fall dormancy were detected between genotypes within the progeny and parental groups. Differences in freeze response were also detected using the artificial freeze treatment. Broad-sense heritabilities were estimated to be 0.895 and 0.573 for spring green-up and fall dormancy respectively. Narrow-sense heritability estimates were found to be 0.885 and 0.265 for these same traits. These results indicate that this population could be improved for cold tolerance using phenotypic recurrent selection. Freeze response was found to be positively correlated to winter hardiness and seed yield in the parent group.

Regression models with correlated errors lead to nonadditivity of the information matrix. This makes the usual approach of design optimization (approximation with a continuous design, application of an equivalence theorem, numerical calculations by a gradient algorithm) impossible. A method is presented that allows the construction of a gradient algorithm by altering the information matrices through adding of supplementary noise. A heuristic is formulated to circumvent the nonconvexity problem and the method is applied to typical examples from the literature. (author's abstract) / Series: Forschungsberichte / Institut für Statistik